Abstract
Background
The impact of specific baseline comorbid conditions on the relative risk of postoperative mortality and periprosthetic joint infection (PJI) in elderly patients undergoing TKA has not been well defined.
Questions/purposes
We calculated the relative risk of postoperative mortality and PJI associated with 29 comorbid conditions in Medicare patients undergoing TKA.
Patients and Methods
The Medicare 5% sample was used to calculate the relative risk of 90-day postoperative mortality and PJI as a function of 29 preexisting comorbid conditions in 83,011 patients who underwent primary TKA between 1998 and 2007.
Results
The independent risk factors for 90-day postoperative mortality (in decreasing order of significance) were congestive heart failure, metastatic cancer, renal disease, peripheral vascular disease, cerebrovascular disease, lymphoma, cardiac arrhythmia, dementia, pulmonary circulation disorders, and chronic liver disease. The independent risk factors for PJI (in decreasing order of significance) were congestive heart failure, chronic pulmonary disease, preoperative anemia, diabetes, depression, renal disease, pulmonary circulation disorders, obesity, rheumatologic disease, psychoses, metastatic tumor, peripheral vascular disease, and valvular disease.
Conclusions
We believe this information important when counseling elderly patients regarding the risks of mortality and PJI after TKA and risk-adjusting publicly reported TKA patient outcomes.
Level of Evidence
Level II, prognostic study. See the Guidelines for Authors for a complete description of levels of evidence.
Introduction
Despite the success of TKA in terms of alleviating pain and improving function in patients with disabling arthritis of the knee, devastating complications such as periprosthetic joint infection (PJI) and death can and do occur [8]. Multiple clinical risk stratification classification systems, including the Charlson CoMorbidity Index [4], the American Society of Anesthesiologists physical status classification (ASA Class) [1], and the All Patient Revised-Diagnosis Related Group Severity of Illness and Risk of Mortality [10], are reliable predictors of morbidity and mortality in certain surgical patients. However, none of these classification systems has been validated in patients undergoing TKA. Furthermore, most surgeons and patients are not familiar with them, and therefore, they are not helpful for preoperative counseling regarding the risks of death and PJI after TKA.
Kurtz et al. [8] previously used the Medicare administrative claims data set to study the risk of PJI after TKA in Medicare patients. This study identified the presence of preexisting patient comorbidities, as defined by the Charlson CoMorbidity index, as a risk factor for PJI along with other factors such as longer procedure duration, lower socioeconomic status, and male gender. However, although the Charlson CoMorbidity Index provides a useful surrogate for the overall health status of surgical patients based on a composite score from 19 conditions, it is not helpful in elucidating the impact of specific diseases on patient outcomes, because patients with different combinations of preexisting conditions may have similar Charlson scores. As a result, the impact of specific baseline comorbid conditions on the relative risk of postoperative mortality and PJI, particularly in the elderly TKA population, has not been well defined. We therefore evaluated the impact of specific baseline comorbid conditions on the relative risk of postoperative mortality and PJI in Medicare patients undergoing TKA.
Patients and Methods
We used the 5% national sample of the Medicare database to evaluate the association between baseline medical comorbidities and the relative risk of 90-day postoperative mortality and PJI in 83,011 patients who underwent primary TKA between 1998 and 2007 with at least 1 year of enrollment before the surgery. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code 81.54 and Current Procedural Terminology, Fourth Edition, code 27447 were used to identify patients undergoing unilateral primary TKA. This study excluded patients who were younger than 65 years old or health maintenance organization enrollees. We excluded health maintenance organization enrollees because their expenditures are not submitted to the Centers for Medicare & Medicaid Services for processing and, therefore, claims from these beneficiaries were not available or may be incomplete.
Using each patient’s unique encrypted Medicare beneficiary identifier, the patients were followed longitudinally throughout the 10-year study period. We tracked the patient’s enrollment status and 90-day postoperative mortality using a linked “denominator” file that accompanied the analytic data sets. The annual Medicare denominator file contains information regarding the date of death, if applicable, of the enrollees and was used to determine the prevalence of 90-day postoperative mortality. Any PJI that was diagnosed during the time period under study (ie, up to the latest followup, December 31, 2007) was identified with ICD-9-CM diagnosis code 996.66 (infection resulting from an internal joint prosthesis) from services provided in any setting, including inpatient, outpatient, office, skilled nursing facility, hospice care, and home health agencies. This ICD-9-CM code is reportedly associated with a high degree of specificity and concordance with the corresponding clinical diagnosis of PJI in the medical record [3]. Preoperative comorbid conditions were compiled from diagnoses in either Part A (inpatient) or Part B (outpatient) claims submitted during the 12-month period before the operation. We included only patients who were enrolled during the entire 12-month period in the study so that a full year of baseline comorbidities could be observed. To minimize misclassification of postoperative complications (eg, postoperative anemia) as preexisting comorbid conditions, we only included those comorbid conditions that were identified in the administrative claims records at least 30 days before surgery.
We used multivariate Cox regression to evaluate the association between the 29 comorbid conditions (Table 1) and 90-day postoperative mortality and PJI. The analysis controlled for age, gender, race, Census region, receipt of public assistance (identified by Medicare buy status for patients whose Medicare premiums and deductibles were subsidized by the state as a result of their financial status), and all other baseline comorbidities. These comorbid conditions were based on the specific diseases that are used to determine the composite Charlson CoMorbidity Index [4] as well as other diseases that are used as comorbidity measures for administrative databases, which were associated with increases in length of hospital stay, hospital charges, complications, and mortality [5]. In addition, preexisting diseases that have been identified in clinical studies as risk factors for PJI were also included [17]. We calculated both the crude relative risk and adjusted hazard ratio associated with each comorbid condition. The p value associated with the test of significance (Wald’s chi square statistic) for the hazard ratio was used to rank the strength of the association of each comorbid condition with 90-day postoperative mortality or PJI while controlling for the other comorbid conditions and other patient factors, as noted previously. The corresponding p values associated with the test statistics for the hazard ratio indicated the relative degree of association or significance of the presence of that specific condition to the outcomes of interest (eg, postoperative mortality or PJI).
Table 1.
Comorbid conditions included in the analysis of risk factors for PJI and 90-day postoperative mortality after TKA
| Comorbid conditions | Disease prevalence |
|---|---|
| Hypertension | 70% |
| Ischemic heart disease | 27% |
| Hypercholesterolemia | 25% |
| Diabetes | 22% |
| History of malignancy | 20% |
| Cardiac arrhythmia | 19% |
| Chronic pulmonary disease | 18% |
| Hypothyroidism | 18% |
| Urinary tract infection | 17% |
| Valvular disease | 12% |
| Peripheral vascular disease | 12% |
| Cerebrovascular disease | 10% |
| Congestive heart failure | 10% |
| Preoperative anemia | 9% |
| Rheumatologic disease | 7% |
| Depression | 7% |
| Obesity | 6% |
| Coagulopathy | 4% |
| Psychoses | 3% |
| Renal disease | 3% |
| Chronic liver disease | 2% |
| Pulmonary circulation | 1% |
| Peptic ulcer disease | 1% |
| Dementia | 1% |
| Lymphoma | 1% |
| Metastatic tumor | 1% |
| Alcohol abuse | 0% |
| Hemiplegia/paraplegia | 0% |
| Drug abuse | 0% |
PJI = prosthetic joint infection.
Results
The results of our analyses demonstrated 10 conditions that are independently associated with an increase in the risk of 90-day postoperative mortality (Table 2). In decreasing order of significance (p < 0.005 for all comparisons), the risk factors for postoperative mortality after TKA in Medicare patients were congestive heart failure (hazard ratio [HR], 2.15; 95% confidence interval [CI], 1.71–2.69), metastatic cancer (HR, 4.40; CI, 2.67–7.26), renal disease (HR, 2.23; CI, 1.68–2.96), peripheral vascular disease (HR, 1.49; CI, 1.20–1.87), cerebrovascular disease (HR, 1.49; CI, 1.19–1.87), lymphoma (HR, 2.20; CI, 1.22–4.0), cardiac arrhythmia (HR, 1.26; CI, 1.04–1.54), dementia (HR, 1.84; CI, 1.10–3.07), pulmonary circulation disorders (HR, 1.72; CI, 1.07–2.76), and chronic liver disease (HR, 1.50; CI, 1.01–2.21). Hypercholesterolemia was associated with a decreased risk of postoperative mortality (HR, 0.64; CI, 0.50–0.80).
Table 2.
Results of multivariable Cox regression analysis to evaluate the independent risk factors for 90-day postoperative mortality in Medicare patients undergoing TKA
| Comorbid conditions | Risk of 90-day postoperative mortality with condition | Risk of 90-day postoperative mortality without condition | Crude relative risk (95% CI) | Adjusted hazard ratio (95% CI) | Wald’s chi-square | p |
|---|---|---|---|---|---|---|
| Congestive heart failure* | 1.93% | 0.49% | 3.94 (3.27–4.75) | 2.15 (1.71–2.69) | 43.64 | <0.0001 |
| Metastatic cancer* | 3.43% | 0.61% | 5.64 (3.60–8.86) | 4.40 (2.67–7.26) | 33.61 | <0.0001 |
| Renal disease* | 2.20% | 0.58% | 3.80 (2.87–5.03) | 2.23 (1.68–2.96) | 30.82 | <0.0001 |
| Hypercholesterolemia† | 0.45% | 0.69% | 0.65 (0.52–0.82) | 0.64 (0.50–0.80) | 14.14 | 0.0002 |
| Peripheral vascular disease* | 1.30% | 0.54% | 2.42 (1.98–2.95) | 1.49 (1.20–1.87) | 12.63 | 0.0004 |
| Cerebrovascular disease* | 1.31% | 0.55% | 2.39 (1.95–2.94) | 1.49 (1.19–1.87) | 12.02 | 0.0005 |
| Lymphoma* | 1.76% | 0.62% | 2.84 (1.57–5.14) | 2.20 (1.22–4.0) | 6.78 | 0.0092 |
| Cardiac arrhythmia* | 1.12% | 0.51% | 2.19 (1.82–2.62) | 1.26 (1.04–1.54) | 5.39 | 0.0203 |
| Dementia* | 2.44% | 0.61% | 3.98 (2.43–6.51) | 1.84 (1.10–3.07) | 5.36 | 0.0206 |
| Pulmonary circulation disorders* | 2.11% | 0.61% | 3.47 (2.27–5.29) | 1.72 (1.07–2.76) | 5.07 | 0.0243 |
| Chronic liver disease* | 1.20% | 0.61% | 1.96 (1.29–2.96) | 1.50 (1.01–2.21) | 4.14 | 0.0420 |
| Chronic pulmonary disease | 0.95% | 0.56% | 1.70 (1.41–2.07) | 1.18 (0.97–1.43) | 2.75 | 0.0973 |
| Diabetes | 0.79% | 0.58% | 1.37 (1.14–1.66) | 1.17 (0.96–1.42) | 2.35 | 0.1249 |
| Ischemic heart disease | 0.97% | 0.50% | 1.95 (1.64–2.32) | 1.16 (0.96–1.39) | 2.29 | 0.1304 |
| Depression | 0.87% | 0.61% | 1.43 (1.06–1.91) | 1.25 (0.91–1.72) | 1.94 | 0.1639 |
| Anemia | 1.03% | 0.59% | 1.76 (1.38–2.24) | 1.17 (0.90–1.53) | 1.32 | 0.2503 |
| Peptic ulcer disease | 0.75% | 0.63% | 1.21 (0.54–2.69) | 0.65 (0.29–1.47) | 1.07 | 0.3002 |
| Alcohol abuse | 1.66% | 0.62% | 2.67 (1.11–6.39) | 1.43 (0.56–3.65) | 0.56 | 0.4549 |
| Urinary tract infection | 0.80% | 0.59% | 1.36 (1.11–1.67) | 1.07 (0.87–1.32) | 0.40 | 0.5266 |
| Obesity | 0.56% | 0.63% | 0.88 (0.61–1.28) | 0.89 (0.60–1.31) | 0.36 | 0.5464 |
| Valvular disease | 1.00% | 0.57% | 1.73 (1.40–2.15) | 0.93 (0.74–1.18) | 0.33 | 0.5654 |
| Rheumatologic disease | 0.67% | 0.62% | 1.08 (0.78–1.49) | 0.92 (0.66–1.28) | 0.27 | 0.6065 |
| Hypertension | 0.66% | 0.55% | 1.21 (0.99–1.47) | 0.95 (0.78–1.16) | 0.22 | 0.6374 |
| Drug abuse | 0.62% | 0.63% | 0.99 (0.14–7.01) | 0.64 (0.09–4.70) | 0.20 | 0.6584 |
| Coagulopathy | 1.05% | 0.61% | 1.73 (1.23–2.43) | 1.07 (0.76–1.52) | 0.17 | 0.6843 |
| Hypothyroidism | 0.63% | 0.63% | 1.00 (0.80–1.26) | 0.97 (0.77–1.23) | 0.06 | 0.8073 |
| Malignancy | 0.82% | 0.58% | 1.40 (1.15–1.71) | 0.98 (0.79–1.22) | 0.02 | 0.8773 |
| Psychoses | 0.93% | 0.62% | 1.51 (1.02–2.23) | 0.98 (0.61–1.59) | 0.01 | 0.9411 |
| Hemiplegia/paraplegia | 1.36% | 0.62% | 2.18 (0.82–5.79) | 1.01 (0.37–2.78) | 0.00 | 0.9838 |
* Condition that is statistically significantly associated with an increased adjusted risk of 90-day post operative mortality in Medicare patients undergoing THA; conditions are listed in descending order of the degree of association with 90-day postoperative mortality based on the p value associated with the hazard ratio; †condition that is associated with a decreased risk of 90-day postoperative mortality; PJI = prosthetic joint infection; CI = confidence interval.
Our analyses identified 13 conditions that are independently associated with an increase in the risk of PJI (Table 3). In decreasing order of significance, the independent risk factors for PJI were congestive heart failure (HR, 1.28; 95% CI, 1.13–1.46), chronic pulmonary disease (HR, 1.22; CI, 1.10–1.36), preoperative anemia (HR, 1.26; CI, 1.09–1.45), diabetes (HR, 1.19; CI, 1.06–1.34), depression (HR, 1.28; CI, 1.08–1.51), renal disease (HR, 1.38; CI, 1.11–1.71), pulmonary circulation disorders (HR, 1.42; CI, 1.06–1.91), obesity (HR, 1.22; CI, 1.03–1.44), rheumatologic disease (HR, 1.18; CI, 1.02–1.37), psychoses (HR, 1.26; CI, 1.02–1.57), metastatic tumor (HR, 1.59; CI, 1.03–2.47), peripheral vascular disease (HR, 1.13; CI, 1.01–1.27), and valvular disease (HR, 1.15; CI, 1.01–1.31).
Table 3.
Results of multivariable Cox regression analysis to evaluate independent risk factors for PJI in Medicare patients undergoing TKA
| Comorbid conditions | Risk of PJI with condition | Risk of PJI without condition | Crude relative risk (95% CI) | Adjusted hazard ratio (95% CI) | Wald’s chi-square | p |
|---|---|---|---|---|---|---|
| Congestive heart failure* | 4.27% | 2.60% | 1.64 (1.47–1.84) | 1.28 (1.13–1.46) | 14.97 | <0.0001 |
| Chronic pulmonary disease* | 3.57% | 2.58% | 1.38 (1.26–1.52) | 1.22 (1.10–1.36) | 14.58 | <0.0001 |
| Preoperative anemia* | 3.71% | 2.67% | 1.39 (1.23–1.58) | 1.26 (1.09–1.45) | 10.18 | 0.0014 |
| Diabetes* | 3.32% | 2.60% | 1.28 (1.17–1.40) | 1.19 (1.06–1.34) | 9.17 | 0.0025 |
| Depression* | 3.69% | 2.69% | 1.37 (1.19–1.58) | 1.28 (1.08–1.51) | 8.53 | 0.0035 |
| Renal disease* | 3.99% | 2.72% | 1.46 (1.20–1.79) | 1.38 (1.11–1.71) | 8.68 | 0.0038 |
| Pulmonary circulation disorders* | 4.98% | 2.73% | 1.82 (1.39–2.38) | 1.42 (1.06–1.91) | 5.37 | 0.0205 |
| Obesity* | 3.72% | 2.70% | 1.38 (1.19–1.59) | 1.22 (1.03–1.44) | 5.25 | 0.0219 |
| Rheumatologic disease* | 3.43% | 2.71% | 1.27 (1.10–1.46) | 1.18 (1.02–1.37) | 4.85 | 0.0277 |
| Pyschoses* | 4.03% | 2.72% | 1.48 (1.23–1.79) | 1.26 (1.02–1.57) | 4.54 | 0.0331 |
| Metastatic tumor* | 4.69% | 2.75% | 1.71 (1.17–2.49) | 1.59 (1.03–2.47) | 4.36 | 0.0369 |
| Peripheral vascular disease* | 3.37% | 2.68% | 1.26 (1.12–1.41) | 1.13 (1.01–1.27) | 4.30 | 0.0381 |
| Valvular disease* | 3.48% | 2.66% | 1.31 (1.17–1.47) | 1.15 (1.01–1.31) | 4.26 | 0.0390 |
| Ischemic heart disease | 3.37% | 2.53% | 1.33 (1.22–1.45) | 1.11 (1.00–1.23) | 3.62 | 0.0572 |
| Cardiac arrhythmia | 3.35% | 2.62% | 1.28 (1.16–1.41) | 1.11 (0.99–1.24) | 3.41 | 0.0649 |
| Coagulopathy | 3.76% | 2.72% | 1.38 (1.16–1.65) | 1.16 (0.96–1.41) | 2.26 | 0.1327 |
| Urinary tract infection | 3.08% | 2.59% | 1.14 (1.03–1.27) | 1.09 (0.97–1.21) | 2.08 | 0.1488 |
| Cerebrovascular disease | 2.93% | 2.74% | 1.07 (0.94–1.21) | 0.91 (0.78–1.06) | 1.57 | 0.2108 |
| Lymphoma | 3.99% | 2.75% | 1.45 (0.99–2.14) | 1.34 (0.85–2.11) | 1.53 | 0.2155 |
| Peptic ulcer disease | 4.40% | 2.74% | 1.60 (1.16–2.22) | 1.23 (0.87–1.75) | 1.43 | 0.2321 |
| Malignancy | 2.99% | 2.70% | 1.11 (1.00–1.22) | 1.07 (0.95–1.20) | 1.33 | 0.2484 |
| Hypercholesterolemia | 2.72% | 2.77% | 0.98 (0.89–1.08) | 0.94 (0.85–1.05) | 1.18 | 0.2763 |
| Hemiplegia/paraplegia | 4.42% | 2.75% | 1.61 (0.94–2.74) | 1.32 (0.77–2.27) | 1.00 | 0.3175 |
| Chronic liver disease | 3.54% | 2.74% | 1.29 (1.02–1.64) | 1.08 (0.84–1.39) | 0.40 | 0.5285 |
| Alcohol abuse | 3.99% | 2.76% | 1.45 (0.83–2.52) | 1.11 (0.63–1.97) | 0.13 | 0.7174 |
| Hypothyroidism | 2.72% | 2.77% | 0.98 (0.88–1.09) | 0.98 (0.87–1.10) | 0.11 | 0.7413 |
| Drug abuse | 4.35% | 2.76% | 1.58 (0.76–3.26) | 1.14 (0.49–2.66) | 0.10 | 0.7569 |
| Hypertension | 2.81% | 2.64% | 1.07 (0.97–1.17) | 1.01 (0.92–1.12) | 0.06 | 0.8025 |
| Dementia | 3.35% | 2.76% | 1.22 (0.80–1.84) | 1.03 (0.66–1.61) | 0.02 | 0.8839 |
* Condition that is statistically significantly associated with an increased adjusted risk of periprosthetic joint infection in Medicare patients undergoing TKA; conditions are listed in descending order of the degree of association with PJI based on the hazard ratio’s p value; PJI = prosthetic joint infection; CI = confidence interval.
Discussion
Death and PJI are rare but devastating complications of TKA. Identifying the impact of specific baseline comorbid conditions on the relative risk of postoperative mortality and PJI, especially in elderly patients, is important for informing discussions between patients and their surgeons when considering elective TKA. Our results indicate that certain baseline comorbid conditions, including congestive heart failure, are associated with a substantially increased risk of postoperative mortality and PJI in elderly patients undergoing TKA.
Our study is limited by a number of factors. First, we relied on administrative claims data, which may not always correlate precisely with the clinical record [12], to identify baseline patient comorbidities. However, the prevalence of each comorbid condition (Table 1) is similar to what has been reported in other population-based studies of patients undergoing TKA [7, 18]. Further study is necessary to better understand the correlation between administratively coded and clinically valid comorbidities. Because our findings are based on the elderly Medicare population, it is unclear if our findings are generalizable to patients younger than 65 years of age. In addition, potentially relevant characteristics such as the patient’s body mass index to indicate the degree of obesity is also not available in the earlier years of the Medicare data set. Because we relied on the diagnoses from the claims records, rather than using clinical criteria such as positive culture or abnormal serology, to identify cases involving PJI, the number of patients with PJI may have been overestimated. However, our previous work demonstrated most deep infections recorded in the Medicare data set were diagnosed while the patient was hospitalized, and most were diagnosed by an orthopaedic surgeon or infectious disease specialists [15], which lend support that these infections were adequately identified from the database. Finally, although we have identified specific comorbidities associated with an increased risk of postoperative mortality and PJI, it is unclear whether certain combinations of risk factors (eg, diabetes and congestive heart failure) result in higher than anticipated risk of mortality and PJI. Further study is necessary to better understand the synergistic effect of combinations of comorbid conditions on the risk of postoperative mortality and PJI in elderly patients undergoing TKA.
Previous investigators have attempted to evaluate the risk factors for mortality and PJI after TKA (Table 4) [8]. In a study involving Medicare patients undergoing TKA, Kurtz et al. [8] identified male gender, public assistance for Medicare premiums, and patient comorbidities, based on the Charlson CoMorbidity Index, as risk factors for PJI. However, as noted, the Charlson CoMorbidity index, which is a composite score reflecting the number and severity of comorbid conditions, provides limited clinical applicability. We focused on identifying which specific patient comorbidities are associated with an increased risk of postoperative mortality and PJI to provide a basis for improved communication and clinical decision-making between surgeons and their patients. Parvizi et al. [16] reviewed the records of 22,540 patients who had undergone elective TKA at a single institution between 1969 and 1997 to identify patients who died within 30 days after the procedure. Mortality rates were determined according to age, gender, diagnosis, implant type, and fixation mode. The 30-day mortality rate was higher for patients with preexisting cardiovascular disease and/or pulmonary disease and simultaneous bilateral TKA. Gill et al. [6] prospectively collected data on 3048 patients who underwent primary elective TKA between 1976 and 1996. Fourteen of the 3048 procedures resulted in death within 90 days after surgery. Patients with cardiac comorbidities had a 16 times higher risk of mortality and risk of mortality in patients who were 85 years and older was 14 times higher. Six patients had a history of cardiac disease defined as a previous hospital admission with a diagnosis of myocardial infarction or ischemic heart disease and cardiac failure.
Table 4.
Data from previous studies examining risk factors for PJI and mortality in patients undergoing TKA
| Pulido et al. [17] (2008) (PJI) | Lee et al. [11] (2006) (SSI) | Lai et al. [9] (2007) (PJI) | Parvizi et al. [16] (2001) (mortality) | Gill et al. [6] (2003) (mortality) | Bozic et al. (current study) (PJI) |
Bozic et al. (current study) (mortality) | |
|---|---|---|---|---|---|---|---|
| Obesity | OR: 3.23, 95% CI (1.6–6.5), p = 0.001 |
HR: 1.22, 95% CI (1.03–1.44), p = 0.0219 |
HR: 0.89, 95% CI (0.60–1.31), p = 0.5464 |
||||
| ASA score > 2 | OR: 1.95, 95% CI (1.0–3.7), p = 0.04 |
||||||
| Bilateral TKA | OR: 5.85, 95% CI (2.5–13.9), p < 0.0001 |
||||||
| Allogeneic blood transfusion | OR: 2.11, 95% CI (1.1–3.9), p = 0.02 |
||||||
| Postoperative atrial fibrillation | OR: 6.22, 95% CI (1.4–28.5), p = 0.02 |
||||||
| Postoperative myocardial infarction | OR: 20.4, 95% CI (2.1–199.9), p = 0.009 |
||||||
| Postoperative urinary infection | OR: 5.45, 95% CI (1.0–8.7), p = 0.04 |
||||||
| Longer hospital stay | OR: 1.09, 95% CI (1.0–1.1), p = 0.0003 |
||||||
| Administrative healthcare facility | OR: 6.25, 95% CI (2.27–16.67), p < 0.001 |
||||||
| Chronic obstructive pulmonary disease | OR: 2.19, 95% CI (0.99–4.86), p = 0.05 |
||||||
| Charlson score of ≥ 3 | OR: 11.28, 95% CI (2.78–45.76), p < 0.001 |
||||||
| Inability to bathe independently | OR: 3.62, 95% CI (1.93–6.81), p < 0.001 |
||||||
| Inability to dress independently | OR: 3.34, 95% CI (1.81–6.18), p < 0.0001 |
||||||
| Diabetes | OR: 3.91, 95% CI (1.06–14.44), p = 0.04 |
HR: 1.19, 95% CI (1.06–1.34), p = 0.0025 |
HR: 1.17, 95% CI (0.96–1.42), p = 0.1249 |
||||
| Total number of comorbid conditions | OR: 1.35, 95% CI (0.94–8.00), p = 0.005 |
||||||
| Preexisting cardiovascular and/or pulmonary disease | Prevalence: 43/47 (91.5%), p < 0.002 |
||||||
| Simultaneous bilateral TKA | Prevalence: 13/47 (27.7%), p < 0.002 |
||||||
| Increasing patient age | OR: 13.7, 95% CI (3.0–44.8), p = 0.001 |
||||||
| Cardiovascular comorbidities | OR: 15.9, 95% CI (3.4–143.5), p < 0.001 |
||||||
| Congestive heart failure | HR: 1.28, 95% CI (1.13–1.46), p < 0.0001 |
HR: 2.15, 95% CI (1.71–2.69), p < 0.0001 |
PJI = prosthetic joint infection; SSI = surgical site infection; HR = hazard ratio; OR = odds ratio; CI = confidence interval; ASA = American Society of Anesthesiologists.
Pulido et al. [17] previously evaluated the risk factors associated with PJI after TJA from 4185 patients undergoing TKA from a single institution from January 2001 to April 2006 using a retrospective cohort study design. Higher ASA score, morbid obesity, bilateral arthroplasty, knee arthroplasty, allogeneic transfusion, postoperative atrial fibrillation, myocardial infarction, urinary tract infection, and longer hospitalization were all identified as risk factors associated with the development of PJI within the first year after TJA. Lee et al. [11] used a case-control study design to evaluate risk factors for surgical site infections in elderly patients who underwent orthopaedic surgery at Duke University Medical Center and seven community hospitals in North Carolina and Virginia between 1991 and 2002. In the bivariate analysis, six variables were associated with surgical site infections. These included admission from a healthcare facility, chronic obstructive pulmonary disease, a Charlson score of 3 or greater, the inability to bathe independently, and the inability to dress independently. Finally, in a retrospective case-control study, Lai et al. [9] examined the individual and cumulative effects of various medical comorbidities on the risk of developing PJI after hip or knee arthroplasty in 51 patients with 52 joint infections. Both diabetes mellitus and total number of medical conditions were associated with higher risk of infection.
Although these studies all provide valuable insights into the risk factors for mortality and PJI after TKA, they are limited by inadequate sample sizes (ranging from 51 to 9245 patients) to detect baseline risk factors for such rare outcomes as death and PJI and inclusion of patients from a single or small number of institutions. Our study builds on these previous findings by identifying the impact of specific baseline comorbid conditions on the relative risk of postoperative mortality and PJI in a large, nationally representative cohort of Medicare patients undergoing TKA. Additionally, our finding that hypercholesterolemia may be associated with a decreased risk of postoperative mortality is interesting in light of recent evidence from the Danish Hip Arthroplasty Registry that suggests the use of cholesterol-lowering agents (eg, statins) may be associated with a decreased risk of revision surgery after primary THA [18]. Although previous authors have identified diabetes and rheumatologic disease as predictors of PJI [9], we found that congestive heart failure, chronic pulmonary disease, preoperative anemia, and depression are also associated with an increased risk of PJI. Depression may also be associated with poor nutritional status, and one study suggests patients undergoing major joint arthroplasty who have preoperative anemia are more likely to receive allogeneic blood transfusions [2], which have been associated with an increased risk of postsurgical infection [14].
In summary, we identified the impact of specific baseline comorbid conditions on the relative risk of 90-day postoperative mortality and PJI after TKA in Medicare patients. Recent literature suggests medical management of diabetes may contribute to lowering the risk of complications after total joint arthroplasty procedures [13], and therefore we recommend optimizing the medical management of the conditions identified in this study before considering elective TKA in this elderly patient population. This information is important when counseling elderly patients regarding the risks associated with TKA. Furthermore, these variables should be included in risk adjustment models for public reporting of TKA outcomes.
Acknowledgments
We thank Vanessa Chiu, MPH, Harry E. Rubash, MD, and Thomas P. Vail, MD, for their assistance.
Financial support Financial support was received from the Orthopaedic Research and Education Foundation. One or more of the authors (SK, KO, EL) are employees of Exponent, Inc. One or more of the authors (DJB) receives consulting income and royalties from DePuy, Inc.
Footnotes
Each author certifies that his or her institution approved the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research.
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.
This work was performed at the University of California, San Francisco and Exponent, Inc.
References
- 1.American Society of Anesthesiologists. ASA Physical Status Classification System. 2010. Available at: http://www.asahq.org/clinical/physicalstatus.htm. Accessed June 25, 2010.
- 2.Borghi 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;17:411–417. doi: 10.1046/j.1365-2346.2000.00693.x. [DOI] [PubMed] [Google Scholar]
- 3.Bozic KJ, Chiu VW, Takemoto SK, Greenbaum JN, Smith TM, Jerabek SA, Berry DJ. The validity of using administrative claims data in total joint arthroplasty outcomes research. J Arthroplasty. 2010;25:58–61. doi: 10.1016/j.arth.2010.04.006. [DOI] [PubMed] [Google Scholar]
- 4.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–383. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 5.Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36:8–27. doi: 10.1097/00005650-199801000-00004. [DOI] [PubMed] [Google Scholar]
- 6.Gill GS, Mills D, Joshi AB. Mortality following primary total knee arthroplasty. J Bone Joint Surg Am. 2003;85:432–435. doi: 10.2106/00004623-200303000-00005. [DOI] [PubMed] [Google Scholar]
- 7.Jones CA, Voaklander DC, Johnston DW, Suarez-Almazor ME. The effect of age on pain, function, and quality of life after total hip and knee arthroplasty. Arch Intern Med. 2001;161:454–460. doi: 10.1001/archinte.161.3.454. [DOI] [PubMed] [Google Scholar]
- 8.Kurtz 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;468:52–56. doi: 10.1007/s11999-009-1013-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.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]
- 10.Lavernia CJ, Laoruengthana A, Contreras JS, Rossi MD. All-Patient Refined Diagnosis-Related Groups in primary arthroplasty. J Arthroplasty. 2009;24:19–23. doi: 10.1016/j.arth.2009.03.008. [DOI] [PubMed] [Google Scholar]
- 11.Lee J, Singletary R, Schmader K, Anderson DJ, Bolognesi M, Kaye KS. Surgical site infection in the elderly following orthopaedic surgery. Risk factors and outcomes. J Bone Joint Surg Am. 2006;88:1705–1712. doi: 10.2106/JBJS.E.01156. [DOI] [PubMed] [Google Scholar]
- 12.Losina E, Barrett J, Baron JA, Katz JN. Accuracy of Medicare claims data for rheumatologic diagnoses in total hip replacement recipients. J Clin Epidemiol. 2003;56:515–519. doi: 10.1016/S0895-4356(03)00056-8. [DOI] [PubMed] [Google Scholar]
- 13.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]
- 14.Marik PE. The hazards of blood transfusion. Br J Hosp Med (Lond) 2009;70:12–15. doi: 10.12968/hmed.2009.70.1.37688. [DOI] [PubMed] [Google Scholar]
- 15.Ong KL, Kurtz SM, Lau E, Bozic KJ, Berry DJ, Parvizi J. Prosthetic joint infection risk after total hip arthroplasty in the Medicare population. J Arthroplasty. 2009;24:105–109. doi: 10.1016/j.arth.2009.04.027. [DOI] [PubMed] [Google Scholar]
- 16.Parvizi J, Sullivan TA, Trousdale RT, Lewallen DG. Thirty-day mortality after total knee arthroplasty. J Bone Joint Surg Am. 2001;83:1157–1161. doi: 10.2106/00004623-200108000-00004. [DOI] [PubMed] [Google Scholar]
- 17.Pulido L, Ghanem E, Joshi A, Purtill JJ, Parvizi J. Periprosthetic joint infection: the incidence, timing, and predisposing factors. Clin Orthop Relat Res. 2008;466:1710–1715. doi: 10.1007/s11999-008-0209-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Thillemann TM, Pedersen AB, Mehnert F, Johnsen SP, Soballe K. The risk of revision after primary total hip arthroplasty among statin users: a nationwide population-based nested case-control study. J Bone Joint Surg Am. 2010;92:1063–1072. doi: 10.2106/JBJS.H.01805. [DOI] [PubMed] [Google Scholar]