Duke Orthopedic Journal

: 2019  |  Volume : 9  |  Issue : 1  |  Page : 43--50

Perioperative blood transfusions and complication rates in total elbow arthroplasty

Swara Bajpai1, Cary S Politzer2, Abiram Bala3, Grant E Garrigues4,  
1 Department of Surgery, University of Alabama, Birmingham, AL, USA
2 Department of Orthopaedic Surgery, University of California San Diego, San Diego, USA
3 Department of Orthopaedic Surgery, Stanford Hospital and Clinics, Redwood City, CA, USA
4 Division of Sports Medicine, Midwest Orthopaedics at Rush, Rush University Medical Center, Chicago, IL, USA

Correspondence Address:
Dr. Cary S Politzer
Department of Orthopaedic Surgery, University of California San Diego, 200 W Arbor Dr, San Diego, California 92003


Aim: Complication rates for total elbow arthroplasty (TEA) have not been well studied. The aim of this study was to analyze its utilization rate, the use of perioperative packed red blood cell transfusions, overall rates of surgical complications, and any association of perioperative transfusions with short-term postoperative medical and long-term surgical complications. Materials and Methods: We queried a large Medicare Standard Analytic Files database from 2005 to 2014 to identify index surgery, perioperative blood transfusions, and postoperative complications. Incidence, odds ratios (ORs), 95% confidence intervals, and P values were calculated for 30-day, 90-day, and 2-year postoperative complications. Results: We identified a total of 7480 primary TEA procedures with a minimum of 2-year follow-up. 577 of these patients received perioperative blood transfusion, whereas 2058 age- and gender-matched controls did not. The transfusion cohort had statistically significantly higher rates of 30-day medical complications such as anemia (OR, 7.54), acute kidney injury (OR, 5.55), sepsis (OR, 4.97), myocardial infarction (OR, 3.36), respiratory failure (OR, 3.17), heart failure (OR, 2.56), pneumonia (OR, 2.46), and urinary tract infection (OR, 2.19). They also had higher rates of infection-related surgical complications within 90 days such as periprosthetic infection (OR, 3.36) and cellulitis/seroma (OR, 2.12) and more blood transfusions (OR, 2.59), which remained significantly elevated at 2 years after surgery. Conclusions: Patients requiring perioperative transfusions after surgery have higher rates of complications. Clinical Significance: Surgeons should preoperatively mitigate transfusion risk and preemptively counsel patients during admission and discharge. Further evidence is needed to determine whether perioperative transfusion is a marker of overall poor health or whether transfusion has an immunomodulatory effect that increases complication risk.

How to cite this article:
Bajpai S, Politzer CS, Bala A, Garrigues GE. Perioperative blood transfusions and complication rates in total elbow arthroplasty.Duke Orthop J 2019;9:43-50

How to cite this URL:
Bajpai S, Politzer CS, Bala A, Garrigues GE. Perioperative blood transfusions and complication rates in total elbow arthroplasty. Duke Orthop J [serial online] 2019 [cited 2022 Dec 7 ];9:43-50
Available from: https://www.dukeorthojournal.com/text.asp?2019/9/1/43/279431

Full Text


Total elbow arthroplasty (TEA) is an effective treatment option for rheumatoid arthritis, osteoarthritis, and severe distal humerus fractures in the elderly.[1] The 5-year survival rate of these implants is up to 90%.[2],[3],[4] In many patients, prosthetic replacement of the elbow can provide pain relief and return of function for patients with severe elbow pathology.[5]

A combination of surgical technique, implant factors, and patients' general medical condition can significantly thwart the long-term outcome of pain relief and functional restoration. Major complications that may require reoperation or revision include deep infection (<1%),[6] symptomatic loosening (incidence 7%–19%),[7],[8] implant dislocation,[6] and periprosthetic fracture (5%–29% in primary TEAs).[7] Neuropraxia, superficial infection, and radiographic loosening may not require further surgery, yet still impact patient satisfaction.[3],[8] Several variables, such as patient comorbidities, poor bone quality, surgical technique, altered surgical field from previous surgeries in the area, patient noncompliance with activity restriction, and trauma may be associated with higher rates of complications.[9]

One factor known to correlate with complications in total hip (THA), knee (TKA), and shoulder (TSA) replacement is the administration of perioperative blood transfusions.[10],[11],[12] Millions of units of packed red blood cells (PRBC) are transfused annually in the United States (US), a significant portion of which are used to treat acute blood-loss anemia perioperatively. Total joint arthroplasties for the hip and knee are among the most frequent procedures requiring transfusions.[11],[13] Despite the apparent high-frequency rates, orthopedic surgeons prefer to limit the use of transfusions due to associated postoperative complications such as sepsis, pneumonia (PNA), and venous thromboembolism.[10],[11],[12]

In the US, there were approximately 615,000 TKA[14] and 27,000 total shoulder arthroplasties performed in 2008,[15] and 310,000 THA in 2010,[16] with the frequency rising over time. This volume dwarfs the experience with TEA, for which there exists no national registry.[17] Thus far, analysis of rate of TEA utilization and complications in the US has been subject to sampling of limited regional or national databases. Recent studies report a small but increasing number of TEAs performed in the US.[6],[17],[18]

The relative infrequency of TEA means that there are limited data on patient outcomes and complications.[4],[7],[17],[19] The key metric for hip and knee arthroplasty in the medicare population is the 90-day prosthetic-related complication rates.[20] Similar guidelines, however, have not been established for TEA. Furthermore, while the use of transfusions during shoulder arthroplasty has been reported,[21] the use of transfusions and associated complications have yet to be reported for TEA.

Our purposes were to determine (1) the annual frequency of TEA procedures performed over 10 years (2005–2014); (2) the prevalence of transfusion; (3) the overall prevalence of surgical complications; and (4) establish trends of perioperative PRBC transfusion with 30-day, 90-day, 2-year, and overall postoperative medical and surgical complications. We hypothesized that the increased incidence of distal humerus fractures in the elderly would lead to an overall increase in TEA utilization. We hypothesized that the transfusion rate would be quite low, given the surgical procedure is generally performed under tourniquet hemostasis. We further hypothesized that overall complication rates would remain low, but that PRBC transfusion would have a significant correlation with higher complication rates.

 Materials and Methods

Utilizing PearlDiver Technologies (Warsaw, Indiana), we queried the Medicare Standard Analytic Files database that contains all inpatient and outpatient facility claims for 51 million covered lives. This database maintains compliance with the Health Insurance Portability and Accountability Act (HIPAA). International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9) and Current Procedural Terminology (CPT) codes were used to identify the surgery, blood transfusions, and postoperative medical and surgical complications of interest [Supplemental Table 1]. Our Institutional Review Board for Clinical Investigations approved the study as exempt research (Pro00050604).{Table 1}[INLINE:1]

To determine overall TEA utilization and transfusion rates over the study period, ICD-9 procedural code 81.84 and CPT code 24,363 were used to identify patients who underwent primary TEA procedures from 2005 to 2014. To avoid confounding effects of multiple surgeries on postoperative complications, we excluded patients who underwent more than one TEA. We then identified all patients who had blood transfusions during the same admission and those who did not.

To determine complication rates, we selected patients with active records for at least 2 years postoperatively from each group to ensure at least 2 years of follow-up. This effectively removed all procedures in 2013 and 2014 and resulted in a transfusion cohort and nontransfusion cohort. We matched the nontransfusion cohort to the transfusion cohort by age and gender proportions and used these final cohorts in all further analyses. Postoperative complications and ICD-9 diagnosis codes were selected based on the experience of arthroplasty surgeons at our own institution and validated codes from the Centers for Medicare and Medicaid Services (CMS) hospital compare data.[22] We identified and compared medical complications within 30-days post-TEA and surgical complications at 90 days and 2 years after surgery for both transfusion and nontransfusion groups.

Statistical methods

For both groups, we calculated incidence, odds ratios (ORs), and 95% confidence intervals (CIs) for each complication. Charlson Comorbidity Index (CCI), a measure of health status, was also calculated for each group. Statistical significance was set to an alpha value of P < 0.05. In compliance with HIPAA standards, queries resulting in between 1 and 10 patients returned a null value and were set to an average of five for the analysis.



We identified 11,672 TEA procedures from 2005 to 2014. As shown in [Figure 1], the utilization of TEA over the study period remained relatively constant, averaging 1101 per year. Blood transfusions peaked in 2008 (11.7%) and 2009 (10.6%) but have since remained relatively stable at 6%–8%, with an average transfusion rate of 8.5%.{Figure 1}

Of the total TEA recipients, 7480 had a minimum 2 years of follow-up with surgery from 2005 to 2012. 577 (7.7%) patients received perioperative blood transfusion and the age- and gender-matched nontransfusion cohort with 2-year follow-up consisted of 2058 patients. These two cohorts are used for all subsequent analyses. About 83.2% of both cohorts were female. The proportion of transfusions nearly doubled from the youngest-to-oldest cohort (12.3% in the <65-year age group to 21.3% in ≥85-year age group). The matching process selectively reduced the proportion of younger male patients in the nontransfusion cohort. The median CCIs for both groups were 5; although, average CCI was higher for the transfusion cohort (6.02 vs. 5.26; mean difference, 0.760; CI, 0.554–0.966; P < 0.01). Therefore, the transfusion group was slightly but significantly sicker on average. Cohort characteristics are shown in [Table 1].

30-day postoperative medical complications

There were a variety of medical complications within 30 days after surgery in patients with and without transfusions [Table 2] and [Figure 2]. The most common short-term medical complications overall were arrhythmia, including atrial fibrillation (16.8%), anemia (14.9%), and urinary tract infection (10.6%). The least frequent complications were stroke (<1%) and (pulmonary embolism [PE]; <1%). Transfused patients had significantly more short-term complications than those who had not received a transfusion except for stroke and PE, which were not different between the two groups. Compared to the nontransfusion group, patients in the transfusion group had greater odds of anemia (OR, 7.54; CI, 5.99–9.50; P < 0.01), acute kidney injury (OR, 5.55; CI, 3.76–8.20; P < 0.01), sepsis/SIRS (OR, 4.97; CI, 2.27–10.9; P < 0.01), myocardial infarction (OR, 3.36; CI, 1.87–6.04; P < 0.01), and respiratory failure (OR, 3.17; CI, 1.54–6.54; P < 0.01), among other medical complications.{Table 2}{Figure 2}

90-day postoperative surgical complications

90-day postoperative surgical complications were infrequent. As shown in [Table 3] and [Figure 3], TEA revision was the most common surgical complication (5.6%), followed by infection-related complications, such as cellulitis/seromas (5.5%) and I and D (incision and drainage) (3.9%). Periprosthetic fracture (0.1%) and loosening (0.1%) were least common. Periprosthetic infections were more frequent in transfused patients (OR, 3.36; CI, 2.12–5.32; P < 0.01), as were bleeding complications (OR, 2.59; CI, 1.49–4.51; P < 0.01), and cellulitis/seromas (OR, 2.12; CI, 1.49–5.32; P < 0.01). Rates of revision and I and D were not different between groups (P = 0.292 and P = 0.31, respectively). Other short-term surgical complications were too rare to assess.{Table 3}{Figure 3}

2-year postoperative surgical complications

For all patients, the most common complication within 2 years after surgery was cellulitis/seroma (11.3%), followed by I and D (7.6%) and TEA revision (6.7%), as shown in [Table 4] and [Figure 4]. Periprosthetic fracture and dislocation were least common (1.9% and 1.7%, respectively). As shown in [Supplemental Table 2], bleeding complications were significantly more frequent in the transfusion group (OR, 4.07; CI, 2.56–6.47; P < 0.01), as well as periprosthetic infections (OR, 1.88; CI, 1.38–2.58; P < 0.01) and cellulitis/seroma (OR, 1.60; CI, 1.23–2.10; P < 0.01). There was statistically no difference in revision rate (P = 0.51), I and D (P = 0.07), or other surgical complications.{Table 4}{Figure 4}[INLINE:2]


The first TEA was performed in 1925[22] and was performed rarely until the linked, “sloppy-hinge” designs of the 1970s lowered problematic dislocation and loosening rates.[23] General postoperative and periprosthetic complications continue to be common following TEA; careful patient selection is required. Postoperative complications of the more common joint arthroplasties (knee and hip) are well defined, including the need for blood transfusions due to intraoperative blood loss.[11] We expected to find an association between perioperative blood transfusions and general medical complications at 30 days after surgery and overall joint-specific complications within 90 days and 2 years of TEA.

We identified 7480 TEA patients in a Medicare database who had the requisite 2-year follow-up. Our hypothesis was that the frequency of TEA would increase over 6 years due to the growing evidence supporting TEA as treatment for distal humerus fractures in the elderly.[2] According to our data, this hypothesis did not hold to be true. It is possible that the simultaneous improvement in the medical management of rheumatoid arthritis has led to an offsetting decrease in TEA utilization for this indication.[24] Consistent with our hypothesis that transfusion rates would be low for TEA given the use of tourniquet hemostasis, transfusion rates varied from 6% to 12% over 8 years. This is substantially <18%–68% usage in THA[11],[13],[25],[26] and 19%–67% in TKA.[11],[13],[26],[27]

Aside from deep vein thrombosis and PEs, 30-day medical complications were more likely in patients that received perioperative transfusions. OR indicate that patients receiving a transfusion were more likely to be diagnosed with anemia, renal failure, or sepsis/SIRS. Although the number of TEA procedures decreased with increasing patient age, the use of perioperative transfusion steadily increased with age. Although it was difficult to determine whether transfusion contributed to the increased risk of complications or if patients requiring transfusion also were at higher risk, care should be taken in patients requiring transfusion for signs of anemia, infection, renal failure, arrhythmia, and heart failure, among others.

Infection is a devastating complication of joint replacement surgery[28] that may require long hospitalizations, multiple surgical procedures, and the possibility of complete explantation of the prosthesis resulting in loss of function, return of pain, and higher costs.[12],[28],[29] Current safe and clean operating procedures, coupled with prophylactic antibiotic therapy, have reduced infection rates from ~7% to ~1%.[29],[30] Certain factors are associated with higher rates of infection after hip, knee, and elbow arthroplasties. For example, patients with rheumatoid arthritis are at 2.6 times greater risk of infection than patients with osteoarthritis.[29] Furthermore, TEA holds a higher infection rate than both TKA (0.6%) and THA (0.9%),[29] probably because the elbow is a subcutaneous joint and it is often the site of skin breakdown and infected bursae.[29] Surprisingly, periprosthetic infection was higher in the transfusion cohort than nontransfusion cohort, whereas I and D typically was not. This could be due to coding that was not captured in I and D such as revision TEA; nonetheless, this should be further explored.

Short- and longer-term surgical complications were infrequent compared to medical complications. Infection-related complications were consistently more common than hardware-related complications, especially in the transfused patient. Previous studies have presented similar associations between allogeneic transfusions and as much as a tenfold increase in rate of infection when compared to autologous transfusions that were unexplained by other risk factors.[12],[26],[31],[32] Allogeneic transfusions were also associated with greater solid tumor recurrence[12],[33] and improved organ transplant survival rates.[12],[33],[34] Thus, transfusion itself may be a risk factor through an immune-modulatory effect that impairs the host's defenses. Further research is needed to delineate exact mechanisms of immunosuppression mediated by allogeneic blood transfusions in humans and how such effects may lead to the medical and surgical complications found in our study.

Overall, the rates of complications seen in this study with TEA are much higher than typical rates for other joint arthroplasties such as THA and TKA. The incidence of cardiac complications and PNA in TKA are reportedly <1%,[27],[35],[36] whereas they were found to range from 1% to 14% in our study. Septicemia rates range from 0.1% during the index admission[27] to 0.3% 90 days following discharge for primary TKA.[37] Our study found a rate of slightly over 1% as early as 30 days postoperatively. Wound infection rates in medicare population-based studies vary between 0.2% and 1%.[35],[37],[38] Periprosthetic joint infection rates within 2 years following THA and TKA range from 1.5% to 2.3%.[39],[40] Our study showed infection rates ranged from 3% to 10% over a period of 2 years after surgery. Incidences for bleeding and hematoma following TKA range from just under 1%[27] to nearly 2%,[36] yet blood transfusions after TEA were found to be over 3% at 2 years. Implant complication rate in TKA is reported to be below 1%[27] but was generally higher in this study.

Our findings should be considered preliminary and not without limitations. Specifically, all studies based on administrative claims are limited by the accuracy and completeness of the coding. For total joint arthroplasty, the comorbidities and complication data in administrative claims databases have been found to be accurate, but incomplete.[41] In addition, while claims database studies give wide-angle views of the overall population, to preserve patient anonymity, details about individual patients, and their particular clinical scenarios are not provided. As we do not have patient-level data, we could not perform a multivariate regression analysis to adjust for all comorbidities, similar to that of other large database studies. However, by matching our groups by age and gender, and finding a similar CCI for both groups, we believe that comorbidities are not entirely responsible for the outcomes. Many patients with rheumatoid arthritis are under age 65 and may not have Medicare insurance and may be underrepresented in this study, as well as younger males in the nontransfusion cohort. Other limitations of this specific database include lack of prescription data to assess perioperative anticoagulation, laboratories to assess perioperative hemoglobin and hematocrit, and surgeon volume and experience level. Furthermore, we cannot assume a cause and effect relationship to explain the differences between the patients who received blood transfusions versus those who did not and randomized trials of blood transfusions postoperative may be difficult to perform for ethical reasons.


The rate of TEA in the US was flat over the study period, yet complications were greater than the complication rates reported for TKA and THA. According to this study, patients with TEA are already at a higher risk of medical complications than other joint implants. Surgeons should be aware that patients requiring postoperative transfusions may have even higher rates of certain complications, particularly, but not limited to, anemia, renal, heart and respiratory failure, other potentially fatal medical problems such as infection-based complications, and bleeding. As such, patients should be preemptively counseled about these risks during admission and at discharge. Clinicians should be especially vigilant for complications when treating TEA patients that require transfusion and should counsel their patients in accordance with their risk profile.

Clinical significance

Administrative database queries are valuable because they contain large patient numbers, particularly for less common procedures like TEA. Reporting outcomes from a nationwide sample provide a stable estimate of common and rare complications in patients who did or did not receive perioperative blood transfusion. Further evidence is needed to determine whether perioperative transfusion is simply a marker of overall poor health or whether the transfusion also has an immune-modulatory effect that increases the risk of certain complications. Surgeons should be especially vigilant for prosthetic-related complications on follow-up of TEA patients that required perioperative transfusions.


The authors acknowledge Dr. Richard Moon for his advice and guidance throughout the project and to Donald T. Kirkendall, ELS, a contracted medical editor, for his assistance in preparing the manuscript for publication.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Gill DR, Morrey BF. The coonrad-morrey total elbow arthroplasty in patients who have rheumatoid arthritis. A ten to fifteen-year follow-up study. J Bone Joint Surg Am 1998;80:1327-35.
2Cobb TK, Morrey BF. Total elbow arthroplasty as primary treatment for distal humeral fractures in elderly patients. J Bone Joint Surg Am 1997;79:826-32.
3Harmer LS, Sanchez-Sotelo J. Total elbow arthroplasty for distal humerus fractures. Hand Clin 2015;31:605-14.
4Plaschke HC, Thillemann TM, Brorson S, Olsen BS. Implant survival after total elbow arthroplasty: A retrospective study of 324 procedures performed from 1980 to 2008. J Shoulder Elbow Surg 2014;23:829-36.
5Welsink CL, Lambers KT, van Deurzen DF, Eygendaal D, van den Bekerom MP. Total elbow arthroplasty: A systematic review. JBJS Rev 2017;5:e4.
6Cook C, Hawkins R, Aldridge JM 3rd, Tolan S, Krupp R, Bolognesi M. Comparison of perioperative complications in patients with and without rheumatoid arthritis who receive total elbow replacement. J Shoulder Elbow Surg 2009;18:21-6.
7Kim JM, Mudgal CS, Konopka JF, Jupiter JB. Complications of total elbow arthroplasty. J Am Acad Orthop Surg 2011;19:328-39.
8Ewald FC, Scheinberg RD, Poss R, Thomas WH, Scott RD, Sledge CB. Capitellocondylar total elbow arthroplasty. J Bone Joint Surg Am 1980;62:1259-63.
9Marsh JP, King GJ. Unlinked total elbow arthroplasty. Instr Course Lect 2015;64:231-42.
10Grier AJ, Bala A, Penrose CT, Seyler TM, Bolognesi MP, Garrigues GE. Analysis of complication rates following perioperative transfusion in shoulder arthroplasty. J Shoulder Elbow Surg 2017;26:1203-9.
11Hart A, Khalil JA, Carli A, Huk O, Zukor D, Antoniou J. Blood transfusion in primary total hip and knee arthroplasty. Incidence, risk factors, and thirty-day complication rates. J Bone Joint Surg Am 2014;96:1945-51.
12Murphy P, Heal JM, Blumberg N. Infection or suspected infection after hip replacement surgery with autologous or homologous blood transfusions. Transfus 1991;31:212-7.
13Bierbaum BE, Callaghan JJ, Galante JO, Rubash HE, Tooms RE, Welch RB. An analysis of blood management in patients having a total hip or knee arthroplasty. J Bone Joint Surg Am 1999;81:2-10.
14Losina E, Thornhill TS, Rome BN, Wright J, Katz JN. The dramatic increase in total knee replacement utilization rates in the United States cannot be fully explained by growth in population size and the obesity epidemic. J Bone Joint Surg Am 2012;94:201-7.
15Kim SH, Wise BL, Zhang Y, Szabo RM. Increasing incidence of shoulder arthroplasty in the United States. J Bone Joint Surg Am 2011;93:2249-54.
16Wolford ML, Palso K, Bercovitz A. Hospitalization for total hip replacement among inpatients aged 45 and over: United States, 2000-2010. NCHS Data Brief 2015;186:1-8.
17Zhou H, Orvets ND, Merlin G, Shaw J, Dines JS, Price MD, et al. Total elbow arthroplasty in the united states: Evaluation of cost, patient demographics, and complication rates. Orthop Rev (Pavia) 2016;8:6113.
18Gay DM, Lyman S, Do H, Hotchkiss RN, Marx RG, Daluiski A. Indications and reoperation rates for total elbow arthroplasty: An analysis of trends in New York state. J Bone Joint Surg Am 2012;94:110-7.
19Lovy AJ, Keswani A, Dowdell J, Koehler S, Kim J, Hausman MR. Outcomes, complications, utilization trends, and risk factors for primary and revision total elbow replacement. J Shoulder Elbow Surg 2016;25:1020-6.
20National Quality Measures Clearinghouse. Total Hip Arthroplasty and/or Total Knee Arthroplasty: hospital-Level Risk-Standardized Complication Rate Following Elective Primary. Total Hip Arthroplasty and/or Total Knee Arthroplasty; 2015. Available from: https://www.qualitymeasures.ahrq.gov/content.aspx?id=49201. [Last accessed on 2016 Jun 23].
21Ryan DJ, Yoshihara H, Yoneoka D, Zuckerman JD. Blood transfusion in primary total shoulder arthroplasty: Incidence, trends, and risk factors in the United States from 2000 to 2009. J Shoulder Elbow Surg 2015;24:760-5.
22Prkić A, van Bergen CJ, The B, Eygendaal D. Total elbow arthroplasty is moving forward: Review on past, present and future. World J Orthop 2016;7:44-9.
23Morrey BF, Bryan RS, Dobyns JH, Linscheid RL. Total elbow arthroplasty. A five-year experience at the Mayo Clinic. J Bone Joint Surg Am 1981;63:1050-63.
24Triplet JJ, Kurowicki J, Momoh E, Law TY, Niedzielak T, Levy JC. Trends in total elbow arthroplasty in the medicare population: A nationwide study of records from 2005 to 2012. J Shoulder Elbow Surg 2016;25:1848-53.
25Pedersen AB, Mehnert F, Overgaard S, Johnsen SP. Allogeneic blood transfusion and prognosis following total hip replacement: A population-based follow up study. BMC Musculoskelet Disord 2009;10:167.
26Rosencher N, Kerkkamp HE, Macheras G, Munuera LM, Menichella G, Barton DM, et al. Orthopedic surgery transfusion hemoglobin european overview (OSTHEO) study: Blood management in elective knee and hip arthroplasty in europe. Transfus 2003;43:459-69.
27Browne JA, Cook C, Hofmann AA, Bolognesi MP. Postoperative morbidity and mortality following total knee arthroplasty with computer navigation. Knee 2010;17:152-6.
28Charnley J. Postoperative infection after total hip replacement with special reference to air contamination in the operating room. Clin Orthop Relat Res 1972;87:167-87.
29Poss R, Thornhill TS, Ewald FC, Thomas WH, Batte NJ, Sledge CB. Factors influencing the incidence and outcome of infection following total joint arthroplasty. Clin Orthop Relat Res 1984;182:117-26.
30Nelson JP, Glassburn AR Jr., Talbott RD, McElhinney JP. The effect of previous surgery, operating room environment, and preventive antibiotics on postoperative infection following total hip arthroplasty. Clin Orthop Relat Res 1980;147:167-9.
31Borghi B, Casati A. Incidence and risk factors for allogenic blood transfusion during major joint replacement using an integrated autotransfusion regimen. The Rizzoli study group on orthopaedic anaesthesia. Eur J Anaesthesiol 2000;17:411-7.
32Vignali A, Braga M, Dionigi P, Radaelli G, Gentilini O, Bellini A, et al. Impact of a programme of autologous blood donation on the incidence of infection in patients with colorectal cancer. Eur J Surg 1995;161:487-92.
33Waymack JP, Balakrishnan K, McNeal N, Gonce S, Miskell P, Warden GD, et al. Effect of blood transfusions on macrophage-lymphocyte interaction in an animal model. Ann Surg 1986;204:681-5.
34Opelz G, Terasaki PI. Improvement of kidney-graft survival with increased numbers of blood transfusions. N Engl J Med 1978;299:799-803.
35Solomon DH, Chibnik LB, Losina E, Huang J, Fossel AH, Husni E, et al. Development of a preliminary index that predicts adverse events after total knee replacement. Arthritis Rheum 2006;54:1536-42.
36Huddleston JI, Maloney WJ, Wang Y, Verzier N, Hunt DR, Herndon JH. Adverse events after total knee arthroplasty: A national medicare study. J Arthroplasty 2009;24:95-100.
37Cram P, Vaughan-Sarrazin MS, Wolf B, Katz JN, Rosenthal GE. A comparison of total hip and knee replacement in specialty and general hospitals. J Bone Joint Surg Am 2007;89:1675-84.
38Mahomed NN, Barrett JA, Katz JN, Phillips CB, Losina E, Lew RA, et al. Rates and outcomes of primary and revision total hip replacement in the United States medicare population. J Bone Joint Surg Am 2003;85:27-32.
39Bongartz T, Halligan CS, Osmon DR, Reinalda MS, Bamlet WR, Crowson CS, et al. Incidence and risk factors of prosthetic joint infection after total hip or knee replacement in patients with rheumatoid arthritis. Arthritis Rheum 2008;59:1713-20.
40Kurtz SM, Ong KL, Lau E, Bozic KJ, Berry D, Parvizi J. Prosthetic joint infection risk after TKA in the medicare population. Clin Orthop Relat Res 2010;468:52-6.
41Bozic KJ, Bashyal RK, Anthony SG, Chiu V, Shulman B, Rubash HE. Is administratively coded comorbidity and complication data in total joint arthroplasty valid? Clin Orthop Relat Res 2013;471:201-5.