Descriptive analysis of total elbow arthroplasty for distal humerus fractures: 30-day complications
Article information
Abstract
Background
Although functional outcomes of total elbow arthroplasty (TEA) for distal humerus fractures are satisfactory, there is a high rate of complications. This study aims to characterize the 30-day complications, readmissions, and mortality of patients with TEA for distal humerus fractures in a large registry database.
Methods
Patients who underwent TEA for a distal humerus fracture were identified from the 2015 to 2020 ACS-NSQIP (American College of Surgeons National Surgical Quality Improvement Program) database. Baseline demographics, clinical characteristics, and complications including deep vein thrombosis/pulmonary embolus, infection, mortality, readmissions, and reoperations were recorded. Overall, 134 patients (mean age, 73.6 years; mean body mass index, 28.9 kg/m2; 88.8% females) were included.
Results
The total complication rate was 21.6% (n=29). The most common complications were unplanned readmission (6.0%), postoperative transfusion (5.2%), unplanned reoperation (3.0%), wound disruption (2.2%), and urinary tract infection (1.5%). The composite infection rate was 3.7%. There were no 30-day events of mortality, sepsis, or cerebral vascular accident. Patients 80 years or older had higher but not significant rates of unplanned readmission (10.2% vs. 3.5%, P=0.116) and reoperation (6.1% vs. 1.2%, P=0.105).
Conclusions
TEA for distal humerus fracture analyzed over 5 years had high rates of 30-day postoperative complications (21.6%), with unplanned readmission, reoperation, and infection being the most common.
Level of evidence
IV
INTRODUCTION
Total elbow arthroplasty (TEA) is a procedure with expanding indications, commonly used for patients with end-stage rheumatoid arthritis, advanced or post-traumatic osteoarthritis of the elbow, and distal humerus fractures [1]. However, our understanding of TEA lags behind that of arthroplasty for other large joints, especially the hip, knee, and shoulder [2]. The risk for complications is much higher compared to hip and knee arthroplasties and warrants further investigation to improve TEA outcomes [3].
In the trauma setting, TEA has been studied as a treatment for distal humerus fractures [4-7]. Prior research has shown high complication rates, and a better understanding of these complications including their associated risk factors is needed. Thus, the purpose of this study is to conduct an analysis of the safety of TEA for distal humerus fractures using a national database with a large sample size and recent data reflecting the advancements in TEA technologies and techniques. We aim to characterize the 30-day complications, mortality, and readmissions following TEA for distal humerus fractures. Additionally, this study aims to identify independent risk factors for reoperation and unplanned readmission for this surgical population.
METHODS
The study was reviewed by the Institutional Review Board of Columbia University and was deemed exempt because it uses publicly available, de-identified data from a national database. Informed consent was waived as the data was de-identified prior to access.
This retrospective study used the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database, covering data from 2015 to 2020. The ACS NSQIP collects high-quality data from more than 700 hospitals nationwide using a robust set of protocols [8-10], all analyses for this study were performed using the IBM SPSS version 28 (IBM Corp.).
Patient Selection
To identify patients who sustained a distal humerus fracture, International Classification of Diseases-Tenth Revision (ICD-10) codes were used. Any patients with an ICD-10 code starting with S42.4 were selected. From this population, only patients who underwent TEA were selected using Current Procedural Terminology (CPT) code 24363 (Arthroplasty, elbow; with distal humerus and proximal ulnar prosthetic replacement e.g., total elbow). These codes are consistent with prior studies [11,12]. Any patient with missing data was excluded from our analysis.
Demographics and Comorbidities
Patient demographics and clinical characteristics were assessed and comprised sex, age, body mass index (BMI), ethnicity, American Society of Anesthesiologists score, principal anesthesia technique, operative time, procedure type (inpatient or outpatient), and triage (non-emergent or emergent). All patients older than 90 years were classified as 90-year-olds for the calculations. Patient comorbidities were also assessed and included diabetes, congestive heart failure, current dialysis, dyspnea, ventilator dependency, history of severe chronic obstructive pulmonary disease (COPD), current smoker, ascites, steroid or immunosuppressant use for a chronic condition, loss of greater than 10% body weight in the last 6 months, bleeding disorders, and transfusion less than 72 hours before surgery.
Postoperative Complications
Thirty-day postoperative complications were assessed and comprised superficial surgical site infection, deep surgical site infection, organ/space infection, wound disruption, pneumonia, unplanned intubation, pulmonary embolism, prolonged ventilator >48 hours, progressive renal insufficiency, urinary tract infection, stroke/cerebrovascular accident, deep vein thrombosis/thrombophlebitis, myocardial infarction, cardiac arrest requiring cardiopulmonary resuscitation, postoperative transfusion, sepsis, unplanned reoperation, and unplanned readmission. A composite surgical infection rate that includes superficial surgical site infection, deep surgical site infection, organ/space infection, and wound disruption was also calculated. Bivariate analyses were performed to compare patient demographics, comorbidities, and clinical characteristics between those who did and did not require reoperation and between those who did and did not have an unplanned readmission. Pearson’s chi-square test was used when appropriate.
RESULTS
Demographics and Comorbidities
A total of 139 patients underwent TEA for treatment of distal humerus fractures. Five of those patients were missing data and excluded from our study, leaving 134 for final analysis (Table 1). Among these patients, the mean age was 73.6 years (standard deviation [SD], 11.3; range, 27–90 years), mean BMI was 28.9 kg/m2 (SD, 6.7 kg/m2), 88.8% (n=119) were females, and 85.1% were White. Forty-nine patients (36.6%) were 80 years of age or older, and 85 patients (63.4%) were younger than 80 years. The BMI distribution was as follows: <18.5 kg/m2 (3.0%, n=4), 18.5–24.9 kg/m2 (26.9%, n=36), 25–29.9 kg/m2 (32.1%, n=43), 30–34.9 kg/m2 (24.6%, n=33), and >35 kg/m2 (13.4%, n=18). The average operative time was 156.4 minutes (SD, 62.3). Procedures in the non-emergent setting constituted 96.3% of the procedures, and 64.9% were inpatient.
Baseline demographics and comorbidities of patients undergoing total elbow arthroplasty for distal humerus fractures
The most common comorbidities were as follows: 26.1% of patients had a diagnosis of diabetes requiring therapy with a non-insulin agent or insulin therapy, 13.4% of the patients were current smokers, 8.2% used steroids or immunosuppressants for a chronic condition, 7.9% had bleeding disorders, and 7.5% of patients had a history of severe COPD (Table 1).
Complications
The rates of postoperative complications are summarized in Table 2. The overall complication rate was 21.6% (n=29). The average total length of hospital stay was 2.9 days (SD, 2.8 days). The five most common complications were unplanned readmission (6.0%), postoperative transfusion (5.2%), unplanned reoperation (3.0%), wound disruption (2.2%), and urinary tract infection (1.5%). There were no events of mortality, sepsis, or stroke/cerebrovascular accident. Regarding surgical site complications, the most common event was wound disruption (n=3, 2.2%). The composite infection rate of urinary tract infection, deep surgical site infection, organ/space infection, and pneumonia was 3.7%. The most common pulmonary complication was pneumonia (n=1, 0.7%). The most common cardiac complication was myocardial infarction (n=1, 0.7%). The incidence of deep vein thrombosis/pulmonary embolism was 0.7% (n=1).
Postoperative complications of patients following total elbow arthroplasty for distal humerus fractures
Unplanned Readmission
Unplanned readmission within 30 days occurred in 10.2% (n=5) of patients aged 80 years or older compared to 3.5% (n=3) of patients aged 79 years or younger (P=0.116). There was no significant difference between those with or without unplanned readmission for history of diabetes (P=0.083), smoking (P=0.322), dyspnea (P=0.848), functional health status (P=0.736), severe COPD (P=0.407), congestive heart failure (P=0.803), or any other baseline demographics or clinical characteristics.
Reoperation
There was a 6.1% (n=3) reoperation rate in patients aged 80 years or older compared to 1.2% (n=1) of patients aged 79 years or younger (P=0.105). All patients who underwent a reoperation had a BMI between 18.5 and 30 kg/m2. There was no difference between those with or without reoperation regarding diabetes (P=0.227), smoking (P=0.491), dyspnea (P=0.923), functional health status (P=0.862), severe COPD (P=0.564), congestive heart failure (P=0.803), or any other baseline demographics or clinical characteristics.
DISCUSSION
The TEA procedure is being increasingly utilized with an expanding list of indications. Although functional outcomes are promising, there remain concerns about associated complications. To minimize patient morbidity and mortality, continual advancements occur in implant technology, implant design, and surgical techniques [13,14]. This study is the first large national database study to evaluate recent data on the morbidity and mortality of patients undergoing TEA for distal humerus fractures. Here, the overall complication rate was 21.6% with an unplanned readmission rate of 6.0% and reoperation rate of 3.0%. No independent risk factors for reoperation or readmission were identified. Although there was an observed trend, age 80 years or older did not significantly increase risk.
Previous studies are limited by sample size or old data that does not accurately reflect recent innovations. Additionally, complication rates in past studies ranged widely, from 9.5% to 52%, highlighting the lack of understanding of the reasons behind these complications [4,6]. The 6.0% unplanned readmission rate observed in this study exceeds rates reported in older studies, which ranged from 0% to 4.2% [15,16]. It is possible that readmission rates may be increasing over time when using TEA for distal humerus fractures, as our study with new data shows higher complication rates relative to previous studies. Patients undergoing more recent TEA may be more surgically and medically complex compared to previous times, although further studies are needed to verify this. Studies that found significantly higher rates of readmission than noted here examined a much longer postoperative period of 90-days and evaluated all TEA indications collectively [17,18].
The 3.0% reoperation rate in this study is similar to or lower than rates in the literature, though much higher rates up to 68% have been documented in studies with longer follow-ups or older data [15,19,20]. Overall, reoperation rates in this and other described studies remain high, and further investigation into risks for reoperation are warranted. This study did not identify any medical comorbidity, baseline demographic, or clinical characteristic that significantly increased the risk for reoperation. However, patients aged 80 years or older did have a tendency for a higher rate of reoperation compared to their younger counterparts.
The 3.7% infection rate reported here, although higher than those of other major joint arthroplasties, is consistent with previously reported TEA infection rates ranging from 3.2% to 11% [21,22]. This higher infection rate may be attributed to the unique anatomical and biomechanical challenges of the elbow joint, which experiences significant stress and limited soft tissue coverage and is more susceptible to infection. Strategies to reduce these risks could involve optimizing surgical techniques, such as improving soft tissue handling and coverage. While these suggestions are speculative, further research is needed to examine these differences relative to other arthroplasty procedures and to develop solutions to address them.
Total rate of complications, readmissions, and reoperations at nearly 22% following TEA for distal humerus fractures is a significant burden to patients and the health care system. A study by Carducci et al. [23] found that TEA has the highest total inpatient cost relative to other joint arthroplasties, including reverse and anatomic total shoulder arthroplasty, total ankle arthroplasty, THA, and TKA [23]. Infection, which occurred at a relatively high rate of 3.7% following TEA in this study, is a life changing complication for patients and typically involves multiple surgeries and reduced functional outcome. Wagner et al. [24] found that a 186% greater median overall hospital cost of $30,338 for a two-stage reimplantation compared to $16,019 for primary TEA. This highlights the need for improved understanding of patient or surgical risk factors for infection to help identify optimal TEA candidates and better strategize preoperative medical optimization.
This study is not without limitations. A large national database is prone to erroneous data entry or data omission. However, stringent data auditing and data collecting training at participating hospitals minimized this risk [10,25,26]. Additionally, the database does not collect surgeon experience with TEA, fellowship training in elbow replacement, or implant manufacturer or system used, which may influence the complication rates. However, use of a multi-institutional database includes a diverse sample of patients and surgeons, which minimizes the influence of these unknown factors. Finally, important data that would improve the interpretation of our findings were not available within the database, such as cause for readmission and reoperation or other variables such as bone mineral density. Further studies examining risk factors for these postoperative complications, reoperations, and readmissions as well as comparison to outcomes of nonoperative care and open reduction and internal fixation are necessary to improve surgical decision making, reduce patient morbidity, and limit costs.
CONCLUSIONS
In conclusion, this descriptive retrospective database study illustrates the safety of TEA for any distal humerus fracture using a large sample size. This study found an overall complication rate approaching 22%, an unplanned readmission rate of 6.0%, a reoperation rate of 3.0%, and a composite infection rate of 3.7% within 30 days postoperatively. Additionally, patients 80 years or older trended toward higher rates of reoperation and readmission compared to their younger counterparts. Orthopedic surgeons should be aware of the high total 30-day risk of TEA for distal humerus fractures to improve patient selection and education.
Notes
Author contributions
Conceptualization: DE, AM, PG, CMJ, DPT. Data curation: DE, AM, CT, PG, JRP, BR. Formal Analysis: DE. Investigation: AM, PG, DPT. Methodology: DE, CT, PG, JRP, BR. Supervision: PG, CMJ, DPT. Writing – original draft: DE, PG. Writing – review & editing: DE, AM, CT, PG, JRP, BR, CMJ, DPT.
Conflict of interest
The authors report the following potential conflicts of interest: CMJ or their immediate family has received financial payments or other benefits from Acumed, Biomet, DePuy, Gotham Surgical, Integra Lifesciences, Journal of the American Academy of Orthopedic Surgeons, Smith & Nephew, Stryker, and Zimmer. This author or a member of their immediate family is a board or committee member of the American Board of Orthopedic Surgery and American Shoulder and Elbow Surgeons. DPT has received grant support from Arthrex and education payments from Arthrex and Smith & Nephew. This research received no specific grant from any funding agency in the public, commercial, or notfor-profit sector.
Funding
None.
Data availability
Contact the corresponding author for data availability.
Acknowledgments
None.