Abstract
Background
This study aims to identify the incidence and risk factors of periprosthetic infections following endoprosthetic reconstruction of femoral metastatic bone disease (MBD). In this population with MBD, the marked impact of infection on the patient’s systemic treatment highlights the importance of understanding both the incidence and associated risk factors.
Methods
This retrospective cohort study included a total of 140 patients who underwent endoprosthetic reconstruction for femoral MBD at a tertiary referral hospital in South Korea between 2009 and 2019. Infection-free survival was estimated using the Kaplan-Meier method, and Cox proportional hazards model analyses were performed to evaluate the risk factors associated with periprosthetic infection.
Results
The incidence of periprosthetic infection in patients who underwent endoprosthetic reconstruction for femoral MBD was 9% (12 out of 140 patients). Risk factors for periprosthetic infection were hepatocellular carcinoma (HCC) as the primary tumor (hazard ratio [HR], 6.08; 95% CI, 1.63–22.6; p = 0.007) and low preoperative absolute neutrophil count (HR, 6.99; 95% CI, 1.79–27.4; p = 0.005).
Conclusions
Patients with femoral MBD had a 9% risk of developing a periprosthetic infection. Given their limited life expectancy, this translated to a substantial rate of 58.9 infections per 1,000 person-joint-years. Possible risk factors for periprosthetic infection were low preoperative absolute neutrophil count and HCC as the primary tumor. The high incidence of periprosthetic infection and its associated risk factors should be considered in patients undergoing endoprosthetic reconstruction for femoral MBD.
Keywords: Secondary bone neoplasms, Periprosthetic infection, Hepatocellular carcinoma, Endoprosthetic reconstruction, Leukocyte count
Periprosthetic infection following surgery for metastatic bone disease (MBD) represents a highly challenging scenario, primarily due to the distinct characteristics of this particular patient group.1) The life expectancy of patients diagnosed with MBD is typically limited, primarily due to the spread of tumors throughout the body. Therefore, continuous systemic therapy is often required to control tumor spread and progression. In the event of periprosthetic infection, there is a significant delay in the systemic treatment of these patients, which is associated with an unfavorable clinical outcome.
A comprehensive understanding of patients with MBD has become increasingly imperative. Cancers of lung, breast, and prostate have great avidity for bone, leading to the development of MBD in many patients. Notably, advances in targeted therapy have prolonged the survival of patients, thereby increasing their susceptibility to MBD. Consequently, the incidence of pathological fractures or impending fractures warranting bone reconstruction surgery has increased among those diagnosed with MBD.2) Endoprosthetic reconstruction is frequently utilized in these patients, resulting in rapid functional restoration and favorable long-term outcomes.3) However, complications such as periprosthetic infection, aseptic loosening, periprosthetic fracture, and mechanical failure are significant risk factors. Specifically, periprosthetic infection increases the risk of repeated surgical interventions, compromised functional outcomes, and adverse prognosis.4)
Periprosthetic infections occurring in patients with MBD exhibit distinct characteristics compared to patients undergoing conventional arthroplasty5,6) or in patients with primary bone tumor.7) It is characterized by systemic tumor burden, diverse histologic tumor types, need for chemotherapy, and limited life expectancy. Compared with those undergoing conventional arthroplasty for osteoarthritis or tumor prosthesis surgery for primary bone tumors, there is a lack of studies specifically investigating periprosthetic infections within this unique patient cohort of MBD, despite their distinct features. Few studies reported the incidence, which ranged from 4.9% to 6.4%, without assessing associated risk factors.8,9,10)
This is the first study to look at risk factors for periprosthetic infections in patients with MBD. We focused on femoral metastases because the femur is the most common site of metastases in the long bones. Our research questions were as follows: What is the incidence of periprosthetic infections in femoral MBD patients following endoprosthetic reconstruction? What are the possible risk factors in these patients?
METHODS
This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. The study was approved by the Institutional Review Board of Seoul National University College of Medicine/Seoul National University Hospital (IRB No. H-2306-149-1441). Informed consent was waived due to the retrospective nature of this study.
Study Design and Setting
In this retrospective cohort study, 140 patients, who underwent primary and revision endoprosthetic reconstruction for femoral MBD at a tertiary hospital in South Korea were included (Fig. 1). All primary and revision surgeries were performed between January 1, 2009 and December 31, 2019. The inclusion criteria were as follows: patients aged 18 years or older, with a histopathologic diagnosis of metastases involving the femur, and those who underwent surgery for bone metastases to the femur. The exclusion criteria were as follows: internal fixation with an intramedullary rod or plate (n = 128), cement-only reconstruction (n = 6), and biologic reconstruction (n = 3).
Fig. 1. A flow diagram illustrating the development of the study cohort of periprosthetic infections in endoprosthetic reconstruction for femoral metastases. A total of 140 patients were evaluated.
Variables and Outcome Measures
Following preoperative patient data were extracted from the tertiary hospital's electronic medical records : age, sex, primary tumor type and location, presence of pathologic or impending fracture, date of operation, death, and the last documented follow-up, visceral, brain, and other skeletal metastases, Eastern Cooperative Oncology Group (ECOG), medical history of diabetes mellitus, presence of chemotherapy or radiation therapy, and preoperative lab findings including hemoglobin (g/dL), absolute neutrophil counts (103/µL), creatinine level (mg/dL), albumin level (g/dL), alkaline phosphatase level (IU/L), sodium level (mg/dL), and calcium level (mg/dL). Primary outcome measure was the diagnosis of periprosthetic infection. Patients who were lost to follow-up were censored at the date of the most recent documented follow-up.
Diagnostic Criteria of Periprosthetic Infection
Periprosthetic infection was diagnosed, based on 1 or more of the following criteria: sinus tract with evidence of communication to the joint, purulence around the prosthesis, synovial fluid with > 1,700 leukocytes/µL or > 65% granulocytes, microbial growth in a preoperative joint aspirate, intraoperative periprosthetic tissue, or sonication fluid in the removed implant. Similar diagnostic criteria for periprosthetic infection were used in studies involving various types of joint prosthesis.11,12,13,14)
Demographics, Description of Study Population
The mean age was 61 years, 46% were women, and 22% of patients had diabetes (Table 1). The most common primary tumor was lung cancer (24%), followed by hepatocellular carcinoma (HCC; 16%), breast cancer (15%), renal cell carcinoma (11%), and prostate cancer (6.4%). Pathologic fracture was present in 66% of the cohort. Forty-five percent of the patients exhibited an ECOG scale of 3 or 4. Visceral and brain metastases were present in 51% and 23%, respectively. Systemic therapy and radiation therapy were administered to 86% and 39% of the cohort, respectively. The 1-year and 2-year survival rates were 41% and 21%, respectively, while the median 1-year survival prediction for this cohort was 30% according to the survival prediction model developed by the Skeletal Oncology Research Group.15)
Table 1. Demographic and Clinical Characteristics of Patients in the Study Cohort.
| Characteristics | Total patients (n = 140) | Infection-free (n = 128) | Periprosthetic infection (n = 12) | p-value | |
|---|---|---|---|---|---|
| Age (yr)* | 61 ± 12 | 62 ± 12 | 59 ± 12 | 0.51 | |
| Sex (female) | 46 (64) | 45 (57) | 58 (7) | 0.38 | |
| Diabetes mellitus | 22 (31) | 23 (30) | 8.3 (1) | 0.30 | |
| Primary tumor type | |||||
| Lung | 24 (33) | 25 (32) | 8.3 (1) | 0.29 | |
| Liver | 16 (22) | 14 (18) | 33 (4) | 0.10 | |
| Breast | 15 (21) | 15 (19) | 17 (2) | 1.00 | |
| Renal | 11 (15) | 12 (15) | 0 | 0.36 | |
| Prostate | 6.4 (9) | 6.3 (8) | 8.3 (1) | 0.56 | |
| Others | 29 (40) | 28 (36) | 33 (4) | 0.74 | |
| Implant type | |||||
| Proximal femoral endoprosthesis | 57 (80) | 57 (73) | 58 (7) | 1.00 | |
| Conventional bipolar hemiarthroplasty | 32 (45) | 34 (43) | 17 (2) | 0.34 | |
| Distal femoral endoprosthesis | 4.2 (6) | 4.9 (6) | 0 | 1.00 | |
| Combined femoral and acetabular reconstruction | 3.6 (5) | 2.3 (3) | 17 (2) | 0.06 | |
| Total femoral endoprosthesis | 2.9 (4) | 2.3 (3) | 8.3 (1) | 0.30 | |
| Operation time† | 180 (152–230) | 180 (150–226) | 195 (163–258) | 0.26 | |
| Revision surgery | 6.4 (9) | 6.3 (8) | 8.3 (1) | 0.57 | |
| Pathologic fracture | 66 (92) | 68 (87) | 42 (5) | 0.11 | |
| ECOG performance scale | |||||
| 1 | 18 (25) | 18 (23) | 17 (2) | 1.00 | |
| 2 | 36 (50) | 38 (48) | 17 (2) | 0.21 | |
| 3 | 40 (56) | 38 (49) | 58 (7) | 0.22 | |
| 4 | 5.0 (7) | 4.9 (6) | 8.3 (1) | 0.47 | |
| Other bone metastases | 78 (109) | 78 (100) | 75 (9) | 0.73 | |
| Visceral metastases | 51 (72) | 50 (64) | 67 (8) | 0.37 | |
| Brain metastases | 23 (32) | 23 (29) | 25 (3) | 1.00 | |
| Chemotherapy | 86 (120) | 87 (111) | 75 (9) | 0.38 | |
| Radiation therapy | 39 (55) | 38 (48) | 58 (7) | 0.22 | |
| Preoperative laboratory findings | |||||
| Hb (g/dL)* | 11.7 ± 1.9 | 11.7 ± 1.9 | 12.2 ± 2.0 | 0.41 | |
| Absolute neutrophil count (103/µL)† | 5.0 (3.7–7.0) | 5.1 (3.7–7.3) | 3.9 (2.5–4.5) | 0.01‡ | |
| Albumin (g/dL)† | 3.6 (3.1–4.0) | 3.6 (3.1–4.0) | 3.8 (3.2–4.0) | 0.35 | |
| Alkaline phosphatase (IU/L)† | 103 (75–176) | 105 (76–183) | 87.5 (63–135) | 0.17 | |
| Calcium (mg/dL)† | 9.0 (8.6–9.4) | 9.0 (8.6–9.4) | 9.0 (8.7–9.2) | 0.70 | |
| Creatinine (mg/dL)† | 0.75 (0.6–1.0) | 0.8 (0.6–1.0) | 0.7 (0.6–0.8) | 0.41 | |
| Sodium (mg/dL)† | 138 (136–141) | 138 (136–141) | 138 (136–141) | 0.87 | |
| 1-Year survival prediction (SORG)† | 0.3 (0.1–0.6) | 0.3 (0.1–0.6) | 0.5 (0.2–0.7) | 0.20 | |
| 1-Year survival | 41 (58) | 40 (51) | 58 (7) | 0.24 | |
| 2-Year survival | 21 (29) | 21 (27) | 17 (2) | 1.00 | |
Values are presented as mean ± standard deviation, percent (number), or median (interquartile range).
ECOG: Eastern Cooperative Oncology Group, SORG: Skeletal Oncology Research Group.
*Assumed normal Gaussian distribution tested by Shapiro-Wilk test. †These variables did not follow normal Gaussian distribution. ‡p < 0.05.
Of the 140 patients who underwent endoprosthetic reconstruction for femoral MBD, the distribution of surgical interventions was as follows: 57% proximal femoral endoprosthesis (n = 80), 32% conventional bipolar hemiarthroplasty (n = 45), 4.2% distal femoral endoprosthesis (n = 6), 3.6% combined femoral and acetabular reconstruction (n = 5), and 2.9% total femoral endoprosthesis (n = 4). The median operation time was 180 minutes (range, 100–450 minutes) and revision surgeries accounted for 6.4% of the procedures.
Statistical Analysis
Baseline characteristics of the cohort were compared between the groups that developed periprosthetic infection and those that did not. Categorical data were analyzed using Fisher’s exact test. For the continuous variables, normality of the data was assessed using the Kolmogorov-Smirnov test or the Shapiro-Wilk test depending on the group size. In cases where the data followed a normal Gaussian distribution, a t-test was used for statistical analysis. Alternatively, if the data did not follow a normal distribution, the Mann-Whitney test was employed. The Kaplan-Meier survival method was performed to estimate the cumulative periprosthetic infection rates. Univariate log-rank test and multivariate Cox proportional hazard regression models were computed to assess the risk factors contributing to periprosthetic infection. The receiver operating characteristic (ROC) curve was generated to evaluate the predictive ability of preoperative absolute neutrophil count (ANC) for periprosthetic infection. The area under the curve (AUC) of the ROC was calculated. A p-value ≤ 0.05 was considered statistically significant, and a 95% CI was reported. IBM SPSS Statistics version 26.0 (IBM Corp.) and R Statistical Software (v4.1.1; R Core Team 2021) were used for statistical analysis.
RESULTS
What is the Incidence of Periprosthetic Infections in NFemoral MBD Patients Following Endoprosthetic Reconstruction?
A total of 140 patients who underwent surgery for MBD involving the femur were evaluated in this study. Among them, 12 patients were diagnosed with periprosthetic infection, resulting in an incidence of 9%. To further assess the incidence of periprosthetic infections adjusting for the short life expectancy, 2 additional measures were utilized.5,6) First, the rate of periprosthetic infection was calculated as 58.9 per 1,000 person-joint-years based on the occurrence of infection in 12 patients over a cumulative follow-up period of 203.6 person-joint-years. Second, the Kaplan-Meier estimator based on censored data (primarily due to patient mortality) was employed to determine the cumulative incidence of periprosthetic infections at 10 years, which was found to be 15.3% (95% CI, 5.2%–24.4%) (Fig. 2). The mean follow-up period for patients following endoprosthetic reconstruction was 18.6 months (95% CI, 14.3–22.9 months), with the average time to development of periprosthetic infection being 6.7 months (95% CI, 2.1–11.2 months).
Fig. 2. The cumulative risk of periprosthetic infections following endoprosthetic reconstruction for metastases to the femur was analyzed using the Kaplan-Meier estimator. At the 5-year time point, the cumulative risk of periprosthetic infections was determined to be 0.153 (95% CI, 0.050–0.243).
The timing of infection was variable. A total of 67% of cases (n = 8) occurred within the first 3 months following surgery, 17% (n = 2) were between 3 and 12 months, and the remaining 17% (n = 2) were after 12 months postoperatively. The causative microorganisms were identified in 9 of the 12 cases examined. The pathogens included coagulase-negative staphylococci (n = 5, 42%), Escherichia coli (n = 1, 8%), Staphylococcus aureus (n = 1, 8%), and Streptococcus agalactiae (n = 1, 8%). One case of polymicrobial infection including both Staphylococcus epidermidis and Enterococcus faecium, was detected in culture. All staphylococcal infections were resistant to methicillin (Table 2).
Table 2. Microbiology of 12 Episodes of Periprosthetic Infection.
| Microbiological characteristics | Episodes | |
|---|---|---|
| Single microorganism | ||
| Coagulase-negative staphylococci* | 42 (5) | |
| Escherichia coli | 8 (1) | |
| Staphylococcus aureus (MRSA) | 8 (1) | |
| Streptococcus agalactiae † | 8 (1) | |
| Polymicrobial† | 8 (1) | |
| No organism identified at initial intraoperative culture | 25 (3) | |
Values are presented as percent (number).
MRSA: methicillin-resistant Staphylococcus aureus.
*Staphylococcus epidermidis (n = 4), Staphylococcus haemolyticus (n = 1). Methicillin-resistance was observed in all 5 isolates of coagulase-negative staphylococci. †Methicillin-resistant Staphylococcus epidermidis and Enterococcus faecalis.
What are the Possible Risk Factors for Periprosthetic Infection in Femoral MBD Patients?
Univariate log-rank and multivariate Cox proportional hazard analyses were performed to evaluate the possible risk factors of periprosthetic infection (Table 3). In the univariate analysis, primary tumor type of HCC (p = 0.004), preoperative ANC < 2,500/µL (p = 0.004), and implant type of combined acetabular and femoral reconstruction (p = 0.02) were identified as risk factors for periprosthetic infection. Sex, revision operation, diabetes mellitus, presence of pathologic fracture, chemotherapy, radiation therapy, operative time, and multiple skeletal metastases were not associated with periprosthetic infection. In multivariate analysis, preoperative ANC < 2,500/µL (hazard ratio [HR], 6.99; 95% CI, 1.79–27.4; p = 0.005) and primary tumor type of HCC (HR, 6.08; 95% CI, 1.63–22.6; p = 0.007) were independent risk factors.
Table 3. Univariate Log-Rank and Multivariate Cox Proportional Hazard Ratio Analyses of the Risk Factors of Periprosthetic Infection after Endoprosthetic Reconstruction for Metastases to the Femur.
| Risk factor | Univariate log-rank | Multivariate | |||||
|---|---|---|---|---|---|---|---|
| n | Infection rate (%) | p-value | HR | 95% CI | p-value | ||
| Age < 50 yr | 24/140 | 17 | 0.10 | ||||
| Female sex | 64/140 | 11 | 0.44 | ||||
| Diabetes mellitus | 31/140 | 3 | 0.19 | ||||
| Preoperative ANC < 2,500/µL | 9/139 | 33 | < 0.01* | 6.99 | 1.79–27.4 | < 0.01* | |
| Primary tumor type | |||||||
| Lung cancer | 33/140 | 3 | 0.25 | ||||
| Hepatocellular carcinoma | 22/140 | 18 | < 0.01* | 6.08 | 1.63–22.6 | < 0.01* | |
| Breast cancer | 21/140 | 10 | 0.86 | ||||
| Renal cell carcinoma | 15/140 | 0 | 0.18 | ||||
| Others | 49/140 | 10 | 0.71 | ||||
| Implant type | |||||||
| Proximal femoral endoprosthesis | 80/140 | 11 | 0.83 | ||||
| Conventional bipolar hemiarthroplasty | 45/140 | 4 | 0.23 | ||||
| Combined femoral and acetabular reconstruction | 5/140 | 40 | 0.02* | ||||
| Total femoral endoprosthesis | 4/140 | 25 | 0.24 | ||||
| Distal femoral endoprosthesis | 6/140 | 0 | 0.42 | ||||
| Pathologic fracture | 92/140 | 5 | 0.17 | ||||
| Chemotherapy | 120/140 | 8 | 0.12 | ||||
| Radiation therapy | 55/140 | 13 | 0.24 | ||||
| Multiple skeletal metastases | 109/140 | 8 | 0.98 | ||||
| Operation time > 3 hr | 72/140 | 10 | 0.68 | ||||
| Revision operation | 9/140 | 22 | 0.20 | ||||
HR: hazard ratio, ANC: absolute neutrophil count.
*p < 0.05.
We further investigated preoperative ANC values that predict prosthetic joint infection, as patients with MBD often suffer pathologic fractures during active chemotherapy, resulting in a low preoperative ANC. ROC curves were drawn and the AUC was calculated to be 0.72 (95% CI, 0.57–0.87; p = 0.014), indicating a moderate discriminatory power for periprosthetic infection. The cutoff value of preoperative ANC < 2,500/µL predicted periprosthetic infection with a sensitivity of 25% and a specificity of 95% (Fig. 3).
Fig. 3. The receiver operating characteristic curve was used to evaluate the predictive ability of absolute neutrophil count (ANC) for periprosthetic infection. The area under the curve was calculated to be 0.716 (95% CI, 0.57–0.87; p = 0.014), indicating moderate discriminatory power. The cutoff value of ANC < 2,500/µL predicted periprosthetic infection with a sensitivity of 25% and a specificity of 95%.
DISCUSSION
Periprosthetic infections can vary significantly depending on factors such as patient characteristics, implant-related factors, and microbiological considerations. Patients with MBD have a shorter life expectancy and carry a systemic tumor burden, which makes a comprehensive understanding of these factors essential when considering patient management. However, there is currently scarce reporting on the incidence and risk factors of periprosthetic infections specifically in MBD patients undergoing endoprosthetic reconstruction. Previous studies have been limited in scope, focusing primarily on a broader group of bone tumors rather than on the specific subgroup of patients with bone metastases. The aim of this study was to investigate the clinical characteristics of periprosthetic infections following endoprosthetic reconstruction for femoral MBD. Specifically, we focused on assessing the incidence of periprosthetic infections and potential contributing factors.
The rate of infection following endoprosthetic reconstruction for femoral MBD was found to be higher compared with conventional arthroplasty and comparable with endoprosthetic reconstruction for primary bone sarcoma. Previously reported periprosthetic infection rates of MBD show considerable variability, ranging from 1.1% to 19%.8,9,10,16,17) In our study, 9% patients were diagnosed with periprosthetic infection. However, this percentage may underestimate the true infection rate due to the relatively short life expectancy of patients with MBD. Indeed, considering the limited life expectancy, we observed a higher periprosthetic infection incidence of 58.9 per 1,000 person-joint-years and a cumulative incidence of 15.3% (95% CI, 5.2%–24.4%) over 10 years. Notably, this infection rate was strikingly higher than that observed for routine hip or knee replacement surgery (1.5 per 1000 person-joint-years).6) The cumulative incidence was also higher than that observed for routine hip or knee arthroplasty (1.4%)6) and comparable to endoprosthetic reconstruction for primary bone tumors (7.2% to 15%).7,12,14,16,18,19) The elevated risk in MBD patients may be attributed to extensive dissections, longer operative times, large endoprosthetic volume, and patient deconditioning due to chemotherapy and radiation, which increases their susceptibility to periprosthetic infections.
In this study, we observed an association between a low preoperative ANC and the occurrence of periprosthetic infection. To our knowledge, no previous study has reported the relationship between low ANC and periprosthetic infection in MBD patients undergoing endoprosthetic reconstruction.4) Our findings indicate an elevated risk of infection in patients with a diminished ANC, a phenomenon frequently observed in individuals with MBD undergoing active chemotherapy. This observation underscores the potential efficacy of preoperative granulocyte-macrophage colony-stimulating factor administration, or delaying surgery post-neutrophil recovery for patients with low ANC. The association between low ANC and infection has been previously reported in abdominal surgery and port insertion procedures.20,21,22) It has also been reported that a lower ANC count in leukemic patients was associated with a higher rate of severe infection.23) The normal range for an ANC is between 2,500 and 6,000, with neutropenia defined as a cutoff of ANC < 1,500/µL. Historically, neutropenia has been considered as a significant risk factor for the development of infection. In this study, a cutoff value of preoperative ANC < 2,500/µL predicted periprosthetic infection with a sensitivity of 25% and a specificity of 95% (Fig. 3). A cutoff value higher than that defining neutropenia was selected to enhance sensitivity while maintaining specificity. Many patients with femoral MBD experience pathologic fractures, which trigger an inflammatory response and recruit neutrophils, thereby elevating their count. Therefore, a higher threshold is necessary for this population to accurately assess infection risk.
Of the 4 most common primary cancers examined in this study, HCC had the highest infection rate at 18.1%. The predisposition of individuals with HCC to increased susceptibility to infection may be due to the frequent occurrence of HCC in a setting of chronic inflammation, including factors such as alcohol use, viral infections, and non-alcoholic fatty liver disease.24,25) Such chronic inflammatory conditions can lead to alterations in both the innate and adaptive immune systems, increasing the susceptibility of individuals to infection. In addition, the coexistence of cirrhosis in patients with HCC, a well-established risk factor for serious bacterial infections, further contributes to the increased susceptibility observed in this population.26)
Of the 9 cases of isolated bacteria, 7 (77.8%) were found to be infected with staphylococci, with 1 case of S. aureus and 6 cases of coagulase-negative staphylococci. This finding is consistent with previous reports indicating that bacterial biofilm formation is often associated with prosthesis infections in tumor patients.4) There was no significant difference in pathogen composition between patients with MBD and those undergoing conventional arthroplasty or primary bone tumor surgery. Staphylococcal infections are known to be associated with biofilm formation. It has been reported that the addition of rifampin to the antibiotic regimen can help to eliminate biofilms produced by staphylococci.27) This approach has been increasingly adopted and may also be a beneficial option for patients with MBD.
Our study has several limitations that should be acknowledged. First, this study is retrospective in nature, which inherently limits the amount of data that can be extracted from medical records. We were unable to use new infection classification systems such as the criteria defined by Parvizi and associates28) or the European Bone and Joint Infection Society definition29) of periprosthetic infection because newer diagnostic methods, such as alpha-defensin test and nuclear imaging, were not performed. However, even without these new criteria, we were able to diagnose using conventional criteria as previously reported.11,12,13,14) As this study was retrospective and conducted at a single center, the generalizability of the findings to broader populations may be limited. Future studies with larger datasets and multicenter cohorts are warranted to provide more robust evidence. Furthermore, transfer bias could be a concern when patients are lost to follow-up, as it is impossible to determine whether they experienced periprosthetic infections and sought care elsewhere. In our cohort, 21 patients were considered lost to follow-up. However, considering the small number of patients in this group, we assume that many of them were lost due to disease progression rather than seeking care at other facilities. Thus, the potential impact of transfer bias is likely minimal. Lastly, the low level of statistical significance observed in some of the analyses may be due to the small number of patients involved. We have refrained from drawing conclusions when no significant differences were observed in the data. However, this study represents the largest cohort available, given the rarity of bone metastases and the specific focus on periprosthetic infections following endoprosthetic reconstruction for femoral metastases.
In conclusion, patients with femoral MBD had 9% risk of developing periprosthetic infection. Given their limited life expectancy, this translated to a substantial rate of 58.9 infections per 1,000 person-joint-years. Possible risk factors for periprosthetic infection were low preoperative ANC and HCC as the primary tumor. The high incidence of periprosthetic infections and their associated risk factors should be considered in patients with femoral MBD.
Footnotes
CONFLICT OF INTEREST: Han-Soo Kim is an Editor-in-Chief of the journal but was not involved in the peer reviewer selection, evaluation, or decision process of this article. No other potential conflicts of interest relevant to this article were reported.
References
- 1.Macedo F, Ladeira K, Pinho F, et al. Bone metastases: an overview. Oncol Rev. 2017;11(1):321. doi: 10.4081/oncol.2017.321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Oster G, Lamerato L, Glass AG, et al. Natural history of skeletal-related events in patients with breast, lung, or prostate cancer and metastases to bone: a 15-year study in two large US health systems. Support Care Cancer. 2013;21(12):3279–3286. doi: 10.1007/s00520-013-1887-3. [DOI] [PubMed] [Google Scholar]
- 3.Errani C, Mavrogenis AF, Cevolani L, et al. Treatment for long bone metastases based on a systematic literature review. Eur J Orthop Surg Traumatol. 2017;27(2):205–211. doi: 10.1007/s00590-016-1857-9. [DOI] [PubMed] [Google Scholar]
- 4.Tsantes AG, Altsitzioglou P, Papadopoulos DV, et al. Infections of tumor prostheses: an updated review on risk factors, microbiology, diagnosis, and treatment strategies. Biology (Basel) 2023;12(2):314. doi: 10.3390/biology12020314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.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(7):1710–1715. doi: 10.1007/s11999-008-0209-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tsaras G, Osmon DR, Mabry T, et al. Incidence, secular trends, and outcomes of prosthetic joint infection: a population-based study, olmsted county, Minnesota, 1969-2007. Infect Control Hosp Epidemiol. 2012;33(12):1207–1212. doi: 10.1086/668421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jeys LM, Grimer RJ, Carter SR, Tillman RM. Periprosthetic infection in patients treated for an orthopaedic oncological condition. J Bone Joint Surg Am. 2005;87(4):842–849. doi: 10.2106/JBJS.C.01222. [DOI] [PubMed] [Google Scholar]
- 8.Chandrasekar CR, Grimer RJ, Carter SR, Tillman RM, Abudu AT. Modular endoprosthetic replacement for metastatic tumours of the proximal femur. J Orthop Surg Res. 2008;3:50. doi: 10.1186/1749-799X-3-50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Potter BK, Chow VE, Adams SC, Letson GD, Temple HT. Endoprosthetic proximal femur replacement: metastatic versus primary tumors. Surg Oncol. 2009;18(4):343–349. doi: 10.1016/j.suronc.2008.08.007. [DOI] [PubMed] [Google Scholar]
- 10.Thorkildsen J, Strom TA, Strom NJ, Sellevold S, Norum OJ. Megaprosthesis for metastatic bone disease-a comparative analysis. Curr Oncol. 2022;29(5):3460–3471. doi: 10.3390/curroncol29050279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Achermann Y, Vogt M, Spormann C, et al. Characteristics and outcome of 27 elbow periprosthetic joint infections: results from a 14-year cohort study of 358 elbow prostheses. Clin Microbiol Infect. 2011;17(3):432–438. doi: 10.1111/j.1469-0691.2010.03243.x. [DOI] [PubMed] [Google Scholar]
- 12.Fujiwara T, Ebihara T, Kitade K, et al. Risk factors of periprosthetic infection in patients with tumor prostheses following resection for musculoskeletal tumor of the lower limb. J Clin Med. 2020;9(10):3133. doi: 10.3390/jcm9103133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Parvizi J, Zmistowski B, Berbari EF, et al. New definition for periprosthetic joint infection: from the Workgroup of the Musculoskeletal Infection Society. Clin Orthop Relat Res. 2011;469(11):2992–2994. doi: 10.1007/s11999-011-2102-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sigmund IK, Gamper J, Weber C, et al. Efficacy of different revision procedures for infected megaprostheses in musculoskeletal tumour surgery of the lower limb. PLoS One. 2018;13(7):e0200304. doi: 10.1371/journal.pone.0200304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Thio QC, Karhade AV, Bindels BJ, et al. Development and internal validation of machine learning algorithms for preoperative survival prediction of extremity metastatic disease. Clin Orthop Relat Res. 2020;478(2):322–333. doi: 10.1097/CORR.0000000000000997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Funovics PT, Hipfl C, Hofstaetter JG, Puchner S, Kotz RI, Dominkus M. Management of septic complications following modular endoprosthetic reconstruction of the proximal femur. Int Orthop. 2011;35(10):1437–1444. doi: 10.1007/s00264-010-1054-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mavrogenis AF, Pala E, Angelini A, et al. Infected prostheses after lower-extremity bone tumor resection: clinical outcomes of 100 patients. Surg Infect (Larchmt) 2015;16(3):267–275. doi: 10.1089/sur.2014.085. [DOI] [PubMed] [Google Scholar]
- 18.Hardes J, Gebert C, Schwappach A, et al. Characteristics and outcome of infections associated with tumor endoprostheses. Arch Orthop Trauma Surg. 2006;126(5):289–296. doi: 10.1007/s00402-005-0009-1. [DOI] [PubMed] [Google Scholar]
- 19.Racano A, Pazionis T, Farrokhyar F, Deheshi B, Ghert M. High infection rate outcomes in long-bone tumor surgery with endoprosthetic reconstruction in adults: a systematic review. Clin Orthop Relat Res. 2013;471(6):2017–2027. doi: 10.1007/s11999-013-2842-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Dogar SA, Khan MA. Implantable port devices in paediatric oncology patients: a clinical experience from a tertiary care hospital. J Pak Med Assoc. 2013;63(10):1248–1251. [PubMed] [Google Scholar]
- 21.Gulack BC, Englum BR, Lo DD, et al. Leukopenia is associated with worse but not prohibitive outcomes following emergent abdominal surgery. J Trauma Acute Care Surg. 2015;79(3):437–443. doi: 10.1097/TA.0000000000000757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Jolissaint JS, Harary M, Saadat LV, et al. Timing and outcomes of abdominal surgery in neutropenic patients. J Gastrointest Surg. 2019;23(4):643–650. doi: 10.1007/s11605-018-04081-0. [DOI] [PubMed] [Google Scholar]
- 23.Bodey GP, Buckley M, Sathe YS, Freireich EJ. Quantitative relationships between circulating leukocytes and infection in patients with acute leukemia. Ann Intern Med. 1966;64(2):328–340. doi: 10.7326/0003-4819-64-2-328. [DOI] [PubMed] [Google Scholar]
- 24.Foerster F, Hess M, Gerhold-Ay A, et al. The immune contexture of hepatocellular carcinoma predicts clinical outcome. Sci Rep. 2018;8(1):5351. doi: 10.1038/s41598-018-21937-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Yang JD, Nakamura I, Roberts LR. The tumor microenvironment in hepatocellular carcinoma: current status and therapeutic targets. Semin Cancer Biol. 2011;21(1):35–43. doi: 10.1016/j.semcancer.2010.10.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gustot T, Durand F, Lebrec D, Vincent JL, Moreau R. Severe sepsis in cirrhosis. Hepatology. 2009;50(6):2022–2033. doi: 10.1002/hep.23264. [DOI] [PubMed] [Google Scholar]
- 27.Zimmerli W, Trampuz A, Ochsner PE. Prosthetic-joint infections. N Engl J Med. 2004;351(16):1645–1654. doi: 10.1056/NEJMra040181. [DOI] [PubMed] [Google Scholar]
- 28.Parvizi J, Tan TL, Goswami K, et al. The 2018 definition of periprosthetic hip and knee infection: an evidence-based and validated criteria. J Arthroplasty. 2018;33(5):1309–1314. doi: 10.1016/j.arth.2018.02.078. [DOI] [PubMed] [Google Scholar]
- 29.McNally M, Sousa R, Wouthuyzen-Bakker M, et al. The EBJIS definition of periprosthetic joint infection. Bone Joint J. 2021;103(1):18–25. doi: 10.1302/0301-620X.103B1.BJJ-2020-1381.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]