Abstract
Introduction
Recent literature, particularly in the fields of total joint arthroplasty, has raised concern for increased perioperative complications following joint arthroplasty in patients with osteoporosis. The purpose of this study is to evaluate the influence of preoperative osteoporosis on 6-month postoperative outcomes following total ankle arthroplasty (TAA).
Methods
The Nationwide Readmissions Database (NRD) was queried from 2015 to 2020 to identify 29,705 patients undergoing primary elective TAA. Patients were divided into two cohorts based on the presence of preoperative osteoporosis, with 943 (3.2 %) having this diagnosis. Preoperative demographics, comorbidities, postoperative outcomes, cost of admission, and total length of stay (LOS) were analyzed between cohorts. Multivariate regression analyses were conducted to control for predictors of adverse postoperative outcomes other than preoperative osteoporosis.
Results
The overall cohort was majority male (54.2 %) with mean age of 64.15 (range 17–90) years, and Charlson-Deyo Comorbidity Index (CCI) score of 0.65 (range 0–12). When stratifying by preoperative osteoporosis, it was found that osteoporosis patients were statistically significantly older (Osteoporosis=68.72 years; Non-osteoporosis=65.03 years; p <.001), more likely to have Medicare insurance ( p <.001), and had a higher CCI score (Osteoporosis=0.98; Non-Osteoporosis=0.64; p <.001). Multivariate regression analysis of 180-day postoperative outcomes found that preoperative osteoporosis was significantly predictive of increased risk of any complication (OR=1.428; p <.001) , transfusion (OR=3.370; p <.001), pulmonary embolism (OR=5.625; p =.016), Pneumonia (OR=3.872; p <.001), non-routine discharge ( OR=1.445; p <.001), and extended stay greater than 4 days (OR=1.310 ; p <.001). Further, preoperative presence of osteoporosis is predictive of nearly a $3000 increase in total cost of procedure (β= 2980.22; p <.001).
Conclusion
The presence of osteoporosis was significantly predictive of higher rates of postoperative complications, LOS, and a substantially higher cost of procedure following TAA. Physicians and patients alike should consider this comorbidity when stratifying risk for outcomes following total ankle arthroplasty.
Level of evidence
Level III, Retrospective Cohort Study
1
Introduction
For patients with end-stage ankle osteoarthritis, total ankle arthroplasty (TAA) and ankle arthrodesis (AA) are the two most common surgical treatments. While the rate of AA has remained relatively constant over the past two decades, the popularity of TAA has drastically increased. , From 2009–2019 the incidence of TAAs more than doubled. This surge in popularity is likely multi-factorial, including better implants/instrumentation, preservation of ankle range of motion, and possibly less risk of adjacent joint arthritis. , Additionally, the evolution of implant design and surgical techniques has improved TAA survivorship, with long-term survival rates ranging from 76 % to 93 %. ,,
Despite advancements in survivorship, mitigating postoperative complications following TAA remains a significant consideration. The most common complications include aseptic loosening, prosthetic joint infection, component subsidence, and periprosthetic fracture. ,, Using the Dutch Arthroplasty Register, Hermus et al. identified risk factors for short-term TAA failure, including higher BMI, younger age, and a history of osteochondritis dissecans treatment.
While osteoporosis is a known risk factor for complications in total hip and total knee arthroplasties, the relationship between osteoporosis and complications following TAA remains less explored. Studies suggest a connection between osteoporosis and poorer outcomes in total hip arthroplasty (THA) and total knee arthroplasty (TKA). For THA, patients with osteoporosis exhibit higher rates of 90-day readmission and higher rate of complications requiring intensive care. Similarly, TKA patients with osteoporosis face higher rates of readmission and surgical complications.
The purpose of this study is to evaluate the influence of preoperative osteoporosis on 6-month postoperative outcomes following TAA. We hypothesize that the presence of preoperative osteoporosis will be associated with a statistically significantly increased risk of postoperative complications, greater length of hospital stay, and higher costs of care following TAA.
2
Methods
The Nationwide Readmissions Database (NRD) was queried from 2015 to 2020 to identify patients undergoing primary elective TAA. Managed by the Healthcare Cost and Utilization Project (HCUP), the NRD is a unique database that provides information regarding national readmissions regardless of patient insurance or age. This dataset contains discharge information for patients with and without repeat visits in a given year, including patients who died in the hospital, providing a weighted estimate of approximately 32 million patients per year. International Classification of Disease 10th revision (ICD-10) Procedure Coding System (PCS) codes were used to identify patients who underwent primary elective TAA (0SRF0JZ, 0SRG0JZ).
Patients identified with concomitant codes for revision TAA (0SWF0JZ, 0SWG0JZ) or removal of TAA (SPF0JZ, 0SPG0JZ) were excluded from the initial patient selection. Additionally, ICD-10 Clinical Modification (CM) codes were used to identify all patients with preoperative diagnosis of osteoporosis with or without current pathological fracture(s) (M81, M80, M80.07XX). The M81 code was used to identify patients with osteoporosis without a current pathological fracture while the M80 indicates osteoporosis with a current pathological fracture. Furthermore, the M80.07 code was used to identify patients that specifically had osteoporosis with a current pathological fracture in the foot and ankle region, denoted by the value “.07”, with subsequent digits “XX” indicative of left or right lower extremities and particular locations such as the metatarsus, tarsus, or toe. Given that the NRD does not track patients across years, patients undergoing surgery in the latter six months of the year were excluded to allow for 180-day follow up. Exclusion of these patients was accounted for utilizing the weighting variable provided by the NRD.
The query ultimately yielded 29,705 patients undergoing primary elective TAA from 2015 to 2020. The overall cohort was then stratified into two cohorts based on the presence of preoperative osteoporosis, with 943 (3.2 %) having this diagnosis. Preoperative demographics and comorbidities, and postoperative outcomes, as well as rates of readmission, reoperations, complications, and mortality within 180-days of operation were compared between the two cohorts. Additionally, cost of admission, and total length of stay (LOS) were analyzed between cohorts.
Preoperatively examined demographics included age, sex, income quartile and primary expected payer. Preoperative comorbidities were examined based on presence of ICD-10 CM code present prior to surgery, with examined variables including obesity, smoking status, history of alcohol abuse, hypertension, diabetes mellitus (DM), congestive heart failure (CHF), peripheral vascular disease (PVD), chronic pulmonary disease, and chronic liver disease. Additionally, Charlson-Deyo Comorbidity Index (CCI), which serves as measure of patients’ risk of 10-year mortality, was compared between groups. CCI is a composite index score that incorporates 17 preoperative comorbidities that are calculated by extraction of ICD-10 CM codes provided by the NRD. Postoperatively, patient outcomes within 180 days including complications, readmissions, reoperations, revisions, and mortality, were examined. Additionally, cost of admission and hospital metrics, including length of stay and discharge disposition, were analyzed for each cohort.
2.1
Statistical analysis
Statistical analysis was performed using IBM SPSS® Statistics software, version 28 (IBM, Armonk, NY, USA). For all statistical tests, an alpha of.05 defined significance. Categorical variables were examined using chi-square tests, and continuous variables were examined via Student t -test. Multivariate regression analyses were conducted to assess predictors of adverse postoperative outcomes, cost of admission, and hospital metrics other than preoperative osteoporosis while controlling for age (years at admission), CCI, indicator of sex, primary expected payer (uniform), obesity, and smoking.
3
Results
The overall cohort was majority male (54.2 %) with mean age of 64.15 (range 17–90) years, and Charlson-Deyo Comorbidity Index (CCI) score of 0.65 (range 0–12). When stratifying by preoperative osteoporosis, it was found that osteoporosis patients were statistically significantly older (Osteoporosis=68.72 years; Non-osteoporosis=65.03 years; p <.001), more likely to have Medicare insurance ( p <.001), and had a significantly higher CCI score (Osteoporosis=0.98; Non-Osteoporosis=0.64; p <.001). Additionally, osteoporosis patients had a statistically significantly higher rate of smoking (Osteoporosis=29.9 %; non-osteoporosis=25.1 %; p =.001) as well as chronic pulmonary disease (Osteoporosis=22.0 %; Non-osteoporosis= 13.0 %; p <.001). However, these patients also had significantly lower rates of obesity (Osteoporosis=16.9 %; Non-osteoporosis= 22.3 %; p <.001) and diabetes mellitus (Osteoporosis=7.21 %; Non-osteoporosis= 11.0 %; p <.001). ( Table 1 )
Table 1
Demographic and Comorbidity Data of Patients Undergoing Total Ankle Arthroplasty (TAA) by History of Osteoporosis.
| Demographic |
Overall
(n = 29,705) |
No Prior osteoporosis
(n = 28,762) |
Prior Osteoporosis (n = 943) | p -value |
|---|---|---|---|---|
| Age, years | 65.14 ± 10.15 | 65.03 ± 10.15 | 68.72 ± 9.51 | <.001* |
| Sex (female) | 45.8 % | 44.4 % | 85.4 % | <.001* |
| Median Household income | .284 | |||
| 1st quartile | 17.8 % | 17.77 % | 16.93 % | |
| 2nd quartile | 27.9 % | 27.84 % | 29.81 % | |
| 3rd quartile | 28.2 % | 28.16 % | 29.28 % | |
| 4th quartile | 26.1 % | 26.21 % | 23.96 % | |
| Insurance | <.001* | |||
| Medicare | 59.2 % | 58.63 % | 77.41 % | |
| Medicaid | 3.4 % | 3.461 % | 3.075 % | |
| Private | 32.8 % | 33.26 % | 19.19 % | |
| Other | 0.29 % | 4.327 % | 0.318 % | |
| Charlson Comorbidity Index (CCI) | 0.65 ± 1.16 | 0.64 ± 1.15 | 0.98 ± 1.60 | <.001* |
| Obese | 22.1 % | 22.3 % | 16.9 % | <.001* |
| Active Smoker | 25.3 % | 25.1 % | 29.9 % | .001* |
| Alcohol Abuse | 1.20 % | 1.22 % | 0.85 % | .364 |
| Hypertension | 58.2 % | 58.2 % | 59.8 % | .314 |
| Diabetes Mellitus | 10.8 % | 11.0 % | 7.21 % | <.001* |
| Congestive Heart Failure (CHF) | 2.37 % | 2.31 % | 4.45 % | <.001* |
| Peripheral Vascular Disease (PVD) | 2.18 % | 2.17 % | 2.75 % | .256 |
| Chronic Pulmonary Disease | 13.3 % | 13.0 % | 22.0 % | <.001* |
| Chronic Liver Desease | 1.44 % | 1.42 % | 2.12 % | .093 |
Following TAA, the rate of complication, readmission, revision, and mortality for the overall cohort was 14.4 %, 9.62 %, 0.19 %, and 0.12 %, respectively. Stratification based on preoperative diagnosis of osteoporosis found statistically significantly higher rates of complication among osteoporosis patients (Osteoporosis=21.4 %; non-osteoporosis= 14.1 %; p <.001), specifically higher perioperative rates of implant associated fracture (Osteoporosis=2.65 %; non-osteoporosis= 1.42 %; p =.002), acute respiratory distress syndrome (ARDS) (Osteoporosis=0.84 %; non-osteoporosis=0.33 %; p =.017), acute renal failure (ARF) (Osteoporosis=1.90 %; non-osteoporosis= 0.73 %; p <.001), other respiratory complications (Osteoporosis=0.32 %; non-osteoporosis=0.04 %; p =.016), transfusion (Osteoporosis=1.17 %; non-osteoporosis=0.23 %; p <.001), pulmonary embolism (Osteoporosis=0.32 %; non-osteoporosis=0.05 %; p =.016), and pneumonia (Osteoporosis=1.06 %; non-osteoporosis=0.12 %; p <.001). However, there was no statistically significant difference in the rate of readmission ( p =.955) , revision ( p =.266), or mortality ( p =.634) between the two cohorts. ( Table 2 )
Table 2
180 Day Postoperative Outcomes of Patients Undergoing TAA by History of Osteoporosis.
|
Overall
(n = 29,706) |
No Prior osteoporosis
(n = 28,762) |
Prior Osteoporosis (n = 943) | p -value | |
|---|---|---|---|---|
| Any Complication | 14.4 % | 14.1 % | 21.4 % | <.001* |
| Implant Associated Fracture | 1.45 % | 1.42 % | 2.65 % | .002* |
| SSI | 0.77 % | 0.78 % | 0.64 % | .567 |
| Wound Dehiscence | 0.17 % | 0.18 % | 0.0 % | .415 |
| ARDS | 0.35 % | 0.33 % | 0.84 % | .017* |
| Sepsis | 0.14 % | 0.14 % | 0.0 % | .642 |
| Urinary Tract Infection | 0.47 % | 0.47 % | 0.85 % | .311 |
| ARF | 0.76 % | 0.73 % | 1.90 % | <.001* |
| Cardiovascular, Including MI | 0.18 % | 0.17 % | 0.53 % | .030 |
| Bleeding | 0.05 % | 0.05 % | 0.0 % | 1.000 |
| GI | 0.02 % | 0.02 % | 0.0 % | 1.000 |
| Peripheral Vascular | 0.02 % | 0.02 % | 0.0 % | 1.000 |
| Shock | 0.01 % | 0.01 % | 0.0 % | 1.000 |
| Respiratory | 0.05 % | 0.04 % | 0.32 % | .016* |
| Transfusion | 0.26 % | 0.23 % | 1.17 % | <.001* |
| Debridement | 0.13 % | 0.14 % | 0.0 % | .642 |
| Cellulitis | 0.12 % | 0.12 % | 0.32 % | .112 |
| Pulmonary Embolism | 0.05 % | 0.05 % | 0.32 % | .016* |
| Pneumonia | 0.15 % | 0.12 % | 1.06 % | <.001* |
| Thrombosis | 0.01 % | 0.02 % | 0.0 % | 1.000 |
| Other | 1.08 % | 1.07 % | 1.59 % | .146 |
| Mortality | 0.12 % | 0.13 % | 0.0 % | .634 |
| Revision | 0.19 % | 0.20 % | 0.0 % | .266 |
| Readmission | 9.62 % | 9.62 % | 9.54 % | .955 |
Related posts:
Clinical and biomechanical outcomes of orthotic devices for progressive collapsing foot deformity: a systematic review and meta-analysis
24: Supracondylar Femur Fractures
The Team Physician
American Academy of Orthopaedic Surgeons Guidelines
Subspine Impingement and Surgical Technique
Biomechanics of Osteosynthesis by Screwed Plates
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
