Abstract
Background
Despite ample evidence supporting cemented femoral fixation for both hemiarthroplasty and THA for surgical treatment of displaced femoral neck fractures, cementless fixation is the preferred fixation method in the United States. To our knowledge, no nationally representative registry from the United States has compared revision rates by fixation for this surgical treatment.
Question/purpose
After controlling for relevant confounding variables, is femoral fixation method (cemented or cementless) in hemiarthroplasty or THA for femoral neck fracture associated with a greater risk of (1) all-cause revision or (2) revision for periprosthetic fracture?
Methods
Patients with Medicare insurance who had femoral neck fractures treated with hemiarthroplasty or THA reported in the American Joint Replacement Registry database from 2012 to 2017 and Centers for Medicare and Medicaid Services claims data from 2012 to 2017 were analyzed in this retrospective, large-database study. Of the 37,201 hemiarthroplasties, 42% (15,748) used cemented fixation and 58% (21,453) used cementless fixation. Of the 7732 THAs, 20% (1511) used cemented stem fixation and 80% (6221) used cementless stem fixation. For both the hemiarthroplasty and THA cohorts, most patients were women and had cementless femoral fixation. Early revision was defined as a procedure that occurred less than 90 days from the index procedure. All patients submitted to the registry were included in the analysis. Patient follow-up was limited to the study period. No patients were lost to follow-up. Due to inherent limitations with the registry, we did not compare medical complications, including deaths attributed directly to cemented fixation. A logistic regression model including the index arthroplasty, age, gender, stem fixation method, hospital size, hospital teaching affiliation, and Charlson comorbidity index score was used to determine associations between the index procedure and revision rates.
Results
For the hemiarthroplasty cohort, risk factors for any revision were cementless stem fixation (odds ratio 1.42 [95% confidence interval 1.20 to 1.68]; p < 0.001), younger age (OR 0.96 [95% CI 0.95 to 0.97]; p < 0.001), and higher Charlson comorbidity index (OR 1.06 [95% CI 1.02 to 1.11]; p = 0.004). Risk factors for early revision were cementless stem fixation (OR 1.77 [95% CI 1.43 to 2.20]; p < 0.001), younger age (OR 0.98 [95% CI 0.97 to 0.99]; p < 0.001), and higher Charlson comorbidity index (OR 1.09 [95% CI 1.04 to 1.15]; p < 0.001). Risk factors for revision due to periprosthetic fracture were cementless fixation (OR 6.19 [95% CI 3.08 to 12.42]; p < 0.001) and higher Charlson comorbidity index (OR 1.16 [95% CI 1.06 to 1.28]; p = 0.002). Risk factors for early revision due to periprosthetic fracture were cementless fixation (OR 7.38 [95% CI 3.17 to 17.17]; p < 0.001), major teaching hospital (OR 2.10 [95% CI 1.08 to 4.10]; p = 0.03), and higher Charlson comorbidity index (OR 1.20 [95% CI 1.09 to 1.33]; p < 0.001). For the THA cohort, there were no associations.
Conclusion
These data suggest that cemented fixation should be the preferred technique for most patients with displaced femoral neck fractures treated with hemiarthroplasty. The fact that stem fixation method did not affect revision rates for those patients with displaced femoral neck fractures treated with THA may be due to current practice patterns in the United States.
Level of Evidence
Level III, therapeutic study.
Introduction
Given the frail nature of many patients who sustain displaced femoral neck fractures, avoiding revision surgery is of critical importance for this patient population. There are ample data showing that cemented femoral fixation is associated with fewer revisions [9, 16, 18]. Despite these data, it is well known that cementless femoral fixation is the preferred technique for hip arthroplasty in the United States [3-6, 8, 10, 12-15].
In the United States, the reason for surgeon preference for cementless fixation is unclear. Likely drivers of this preference are ease of use, concerns about cement implantation syndrome, and the success of cementless femoral fixation in patients with hip osteoarthritis treated with THA. Further, surgeons may perceive a lack of applicability of international data to the United States patient population and healthcare system. Perhaps data from a nationally representative registry like the American Joint Replacement Registry (AJRR) will promote change [3-6, 8-10, 12-18].
Therefore, using AJRR data and after controlling for relevant confounding variables, we asked: Is femoral fixation method (cemented or cementless) in hemiarthroplasty or THA for femoral neck fracture associated with a greater risk of (1) all-cause revision or (2) revision for periprosthetic fracture?
Patients and Methods
Study Setting and Data Sources
This retrospective, large-database study drew data from the AJRR, which is a nationwide database containing information on more than 2.1 million procedures representing 1410 institutions, with data from hospitals, ambulatory surgery centers, and private practice groups from all 50 states and the District of Columbia as of October 2020 [2]. The AJRR dataset was merged with the Centers for Medicare and Medicaid Services (CMS) claims dataset for patients 65 years and older. Supplementing the AJRR dataset with the CMS database improved our ability to capture linked revision procedures when the revision procedure was not performed at an AJRR hospital. A recent study comparing AJRR patients with patients in the United States Nationwide Inpatient Sample confirms that AJRR is indeed a nationally representative sample. Linking AJRR data to the CMS claims data allows for capture of 100% of patient data, so there is no “lost to follow-up” [16].
Procedures were linked with CMS claims data from 2012 to 2017 to obtain data on clinical outcomes including revision. We identified the target outcomes of early revision and any revision for a variety of indications. Early revision was defined as revision within 90 days of the index procedure.
Patients and Inclusion/Exclusion Criteria
All primary THA procedures in patients 65 years or older with femoral neck fractures reported to the AJRR from 2012 to 2017 were included. This yielded 37,201 hemiarthroplasties and 7732 THAs. For the hemiarthroplasties, 42% (15,748 of 37,201) used cemented fixation and 58% (21,453 of 37,201) used cementless fixation (Table 1). For the THAs, 20% (1511 of 7732) used cemented stem fixation and 80% (6221 of 7732) used cementless stem fixation (Table 2).
Table 1.
Baseline characteristics and univariate analysis for hemiarthroplasty patients
| Hemiarthroplasty | Total (n = 37,201) | Revision | Early revision | PPFx | Early PPFx | Revised p value | Early revision p value | PPFx p value | Early PPFx p value |
| Stem fixation | < 0.001 | < 0.001 | < 0.001 | < 0.001 | |||||
| Cemented | 42 (15,748) | 1 (221) | 1 (126) | 0.1 (9) | 0 (6) | ||||
| Cementless | 58 (21,453) | 2 (458) | 2 (316) | 0.4 (80) | 0.3 (65) | ||||
| Gender | 0.32 | 0.57 | 0.53 | 0.95 | |||||
| Women | 69 (25,790) | 2 (459) | 1 (301) | 0.2 (59) | 0.2 (49) | ||||
| Men | 31 (11,408) | 2 (220) | 1 (141) | 0.3 (30) | 0.2 (22) | ||||
| Unknown | 0 (3) | ||||||||
| Hospital size | 0.83 | 0.77 | 0.1 | 0.08 | |||||
| Between 1-99 beds | 11 (3946) | 2 (68) | 43 (1.0) | 0.1 (5) | 0.1 (4) | ||||
| Between 100-399 beds | 53 (19,826) | 2 (361) | 1 (242) | 0.3 (57) | 0.2 (47) | ||||
| ≥ 400 Beds | 36 (13,268) | 2 (248) | 1 (156) | 0.2 (27) | 0.2 (20) | ||||
| Unknown | 0.4 (161) | 0 (2) | 0 (1) | ||||||
| Teaching affiliation | 0.76 | 0.51 | 0.18 | 0.13 | |||||
| Major | 16 (5803) | 2 (112) | 1 (75) | 0.3 (20) | 0.3 (17) | ||||
| Nonteaching | 24 (9091) | 2 (169) | 1 (114) | 0.2 (22) | 0.2 (18) | ||||
| Minor | 60 (22,224) | 2 (398) | 1 (253) | 0.2 (47) | 0.2 (36) | ||||
| Unknown | 0 (83) | ||||||||
| Region | 0.18 | 0.18 | 0.08 | 0.09 | |||||
| Northeast | 15 (5509) | 2 (108) | 1.4 (75) | 0.3 (19) | 0.3 (15) | ||||
| Midwest | 30 (10,711) | 2 (179) | 1 (111) | 0.1 (16) | 0.1 (12) | ||||
| South | 29 (10,762) | 2 (217) | 1 (140) | 0.3 (30) | 0.2 (2) | ||||
| West | 27 (10,200) | 2 (175) | 1 (116) | 0.2 (24) | 0.3 (19) | ||||
| Missing | 0.0 (19) | 1 | |||||||
| Age | 83.8 ± 7.9 | 81.0 ± 8.0 | 82.2 ± 7.9 | 82.5 ± 7.9 | 82.1 ± 8.4 | < 0.001 | < 0.001 | 0.121 | 0.07 |
| Charlson comorbidity index | 0.81 ± 1.6 | 0.97 ± 1.6 | 1.0 ± 1.6 | 1.3 ± 1.7 | 1.5 ± 1.8 | 0.008 | 0.001 | 0.003 | 0.001 |
Data presented as % (n) and mean ± SD; PPFx = periprosthetic fracture.
Table 2.
Baseline characteristics and univariate analysis for THA patients
| THA | Total (n = 7732) | Revision | Early revision | PPFx | Early PPFx | Revised p value | Early revision p value | PPFx p value | Early PPFx p value |
| Stem fixation | 0.06 | 0.06 | 0.05 | 0.20 | |||||
| Cemented | 20 (1511) | 2 (30) | 1 (18) | 0.1 (1) | 0.1 (1) | ||||
| Cementless | 80 (6221) | 3 (177) | 2 (118) | 0.4 (24) | 0.3 (16) | ||||
| Gender | 0.69 | 0.40 | 0.94 | 0.87 | |||||
| Women | 69 (5315) | 3 (145) | 2 (98) | 0.3 (17) | 0.2 (12) | ||||
| Men | 31 (2415) | 3 (62) | 2 (38) | 0.3 (8) | 0.2 (5) | ||||
| Unknown | 0 (2) | ||||||||
| Hospital size | < 0.001 | 0.01 | 0.13 | 0.50 | |||||
| Between 1-99 beds | 11 (853) | 1 (19) | 2 (14) | 1 (4) | 0.4 (3) | ||||
| Between 100-399 beds | 48 (3686) | 2 (80) | 1 (50) | 0.2 (7) | 0.2 (6) | ||||
| ≥ 400 beds | 41 (3140) | 3 (108) | 2 (72) | 0.4 (14) | 0.3 (8) | ||||
| Unknown | 1 (53) | ||||||||
| Teaching affiliation | 0.04 | 0.47 | 0.17 | 0.80 | |||||
| Major | 18 (1363) | 3 (47) | 2 (28) | 0.6 (8) | 0.3 (4) | ||||
| Nonteaching | 21 (1636) | 2 (32) | 2 (24) | 0.2 (4) | 0.2 (3) | ||||
| Minor | 61 (4696) | 3 (128) | 2 (84) | 0.3 (13) | 0.2 (10) | ||||
| Unknown | 1 (37) | ||||||||
| Region | 0.51 | 0.43 | 0.61 | 0.73 | |||||
| Northeast | 18 (1387) | 3 (42) | 2 (30) | 0.4 (6) | 0.3 (4) | ||||
| Midwest | 22 (1732) | 2 (38) | 1 (24) | 0.2 (3) | 0.1 (2) | ||||
| South | 30 (2325) | 3 (65) | 2 (42) | 0.3 (8) | 0.3 (6) | ||||
| West | 30 (2280) | 3 (61) | 2 (39) | 0.4 (8) | 0.2 (5) | ||||
| Missing | 0.1 (8) | 0 (1) | |||||||
| Age | 77.8 ± 8.2 | 77.1 ± 7.5 | 77.8 ± 7.9 | 79 ± 7.7 | 77.7 ± 7.7 | 0.24 | 0.94 | 0.47 | 0.99 |
| Charlson comorbidity index | 0.8 ± 1.5 | 0.9 ± 1.5 | 0.9 ± 1.4 | 0.9 ± 1.8 | 0.5 ± 0.7 | 0.62 | 0.58 | 0.91 | 0.09 |
Data presented as % (n) and mean ± SD; PPFx = periprosthetic fracture.
Primary and revision THAs and associated diagnoses were identified using ICD-9, ICD-10, and Current Procedural Terminology codes in both the AJRR and the CMS datasets. Revisions were linked to a primary THA by confirming patient identifiers including Social Security numbers, Medicare beneficiary identifiers, procedure site, and laterality. Laterality was determined by either ICD-10 coding or a discrete AJRR data element when only ICD-9 revision codes were available. Procedures were excluded if laterality could not be confirmed. Hospital size and teaching status were defined by the American Hospital Association Data Survey, Fiscal Year 2015 [1]. The American Hospital Association categorizes hospital teaching status into major teaching hospitals (hospitals with a Council of Teaching Hospitals designation), minor teaching hospitals (hospitals approved to participate in residency and/or internship training by the Accreditation Council for Graduate Medical Education or American Osteopathic Association, or those with the medical school affiliation reported to the American Hospital Association), or nonteaching hospitals (hospitals without Council of Teaching Hospitals, Accreditation Council for Graduate Medical Education, American Osteopathic Association, or medical school affiliations). For this analysis, major and minor teaching hospitals were combined into one teaching group and compared with nonteaching hospitals. The American Hospital Association categorizes hospital size into small hospitals (1-99 beds), medium-sized hospitals (100-399 beds), and large hospitals (≥ 400 beds) [1].
Primary and Secondary Study Outcomes
The primary study outcome was all-cause revision. Revision is defined as any reoperation involving the previously operated hip. The secondary outcomes include revision for periprosthetic fracture as well as early revision. We included early revision in the analysis as it is heavily influenced by surgical technique and carries a high degree of morbidity and mortality. We did not include death in our analysis as it is not included in the AJRR database nor in the CMS claims data. Although it is available from the National Death Index, we felt that it was beyond the scope of our current study. Further, we did not include data on medical complications as these are beyond the scope of this investigation as well.
Ethical Approval
Ethics approval was not obtained for this study as all data are deidentified.
Statistical Analysis
We used a multivariable logistic regression model, informed from results of a univariate analysis, to determine associations between the index procedure and all-cause revision, early revision, and revision for periprosthetic fracture. We controlled for the independent variables of the index arthroplasty, age, gender, stem fixation method, hospital size, hospital teaching affiliation, and Charlson comorbidity index score. Adjustment for multiple comparisons was performed. All analyses were performed by statisticians at the American Academy of Orthopaedic Surgeons using SAS version 9.4 (SAS Institute Inc). Statistical significance was set at p < 0.05.
Results
All-cause Revision
Hemiarthroplasty
After controlling for relevant confounding variables such as stem fixation method, age, and Charlson comorbidity index, we found the risk of all-cause revision after hemiarthroplasty for femoral neck fracture was higher among patients treated with cementless implants compared with those treated with cemented implants (odds ratio 1.42 [95% confidence interval 1.20 to 1.68]; p < 0.001). Other risk factors for any revision included younger age (OR 0.96 [95% CI 0.95 to 0.97]; p < 0.001) and higher Charlson comorbidity index (OR 1.06 [95% CI 1.02 to 1.11]; p = 0.004). Risk factors for any early revision were cementless stem fixation (OR 1.77 [95% CI 1.43 to 2.20]; p < 0.001), younger age (OR 0.98 [95% CI 0.97 to 0.99]; p < 0.001), and a higher Charlson comorbidity index (OR 1.09 [95% CI 1.04 to 1.15]; p < 0.001) (Table 3).
Table 3.
Hemiarthroplasty model for any and early revision
| All-cause, all-time revision | |||
| Effect | Odds ratio | 95% Wald confidence limits | p value |
| Stem fixation: cementless vs cemented | 1.42 | 1.20-1.68 | < 0.001 |
| Gender: women vs men | 0.99 | 0.84-1.17 | 0.89 |
| Size: ≥ 400 beds vs between 100-399 beds | 1.03 | 0.86-1.24 | 0.75 |
| Size: between 1-99 beds vs between 100-399 beds | 0.97 | 0.74-1.27 | 0.83 |
| Teaching hospital: major vs minor | 1.01 | 0.80-1.27 | 0.96 |
| Teaching hospital: non-teaching vs minor | 1.01 | 0.83-1.23 | 0.93 |
| Region: Midwest vs West | 1.03 | 0.83-1.28 | 0.77 |
| Region: Northeast vs West | 1.16 | 0.90-1.50 | 0.25 |
| Region: South vs West | 1.09 | 0.88-1.33 | 0.43 |
| Age | 0.96 | 0.95-0.97 | < 0.001 |
| Charlson comorbidity index | 1.06 | 1.02-1.11 | 0.004 |
| Early revision | |||
| Effect | Odds ratio | 95% Wald confidence limits | p value |
| Stem fixation: cementless vs cemented | 1.77 | 1.43-2.20 | < 0.001 |
| Gender: women vs men | 1.01 | 0.82-1.24 | 0.92 |
| Size: ≥ 400 beds vs between 100-399 beds | 0.96 | 0.76-1.21 | 0.72 |
| Size: between 1-99 beds vs between 100-399 beds | 0.91 | 0.65-1.27 | 0.58 |
| Teaching hospital: major vs minor | 1.09 | 0.82-1.44 | 0.57 |
| Teaching hospital: non-teaching vs minor | 1.03 | 0.81-1.31 | 0.80 |
| Region: Midwest vs West | 0.98 | 0.75-1.28 | 0.87 |
| Region: Northeast vs West | 1.19 | 0.87-1.62 | 0.28 |
| Region: South vs West | 1.04 | 0.80-1.33 | 0.79 |
| Age | 0.98 | 0.97-0.99 | < 0.001 |
| Charlson comorbidity index | 1.09 | 1.04-1.15 | < 0.001 |
THA
After controlling for confounding variables, the risk of all-cause revision after THA for femoral neck fracture was not higher among patients treated with cementless implants compared with those treated with cemented implants (OR 1.34 [95% CI 0.90 to 2.01]; p = 0.15) (Table 4).
Table 4.
THA model for any and early revision
| All-cause, all-time revision | |||
| Effect | Odds ratio | 95% Wald confidence limits | p value |
| Stem fixation: cementless vs cemented | 1.34 | 0.90-2.01 | 0.15 |
| Gender: women vs men | 1.08 | 0.79-1.46 | 0.64 |
| Size: ≥ 400 beds vs between 100-399 beds | 1.43 | 1.02-1.99 | 0.03 |
| Size: between 1-99 beds vs between 100-399 beds | 1.10 | 0.65-1.85 | 0.73 |
| Teaching hospital: major vs minor | 1.11 | 0.77-1.61 | 0.59 |
| Teaching hospital: non-teaching vs minor | 0.80 | 0.52-1.22 | 0.30 |
| Region: Midwest vs West | 0.77 | 0.51-1.17 | 0.21 |
| Region: Northeast vs West | 0.97 | 0.63-1.48 | 0.88 |
| Region: South vs West | 0.90 | 0.62-1.30 | 0.57 |
| Age | 0.99 | 0.98-1.01 | 0.39 |
| Charlson comorbidity index | 1.03 | 0.94-1.13 | 0.52 |
| Early revision | |||
| Effect | Odds ratio | 95% Wald confidence limits | p value |
| Stem fixation: cementless vs cemented | 1.54 | 0.92- 2.57 | 0.10 |
| Gender: women vs men | 1.18 | 0.81-1.73 | 0.39 |
| Size: ≥ 400 beds vs between 100-399 beds | 1.62 | 1.07-2.44 | 0.02 |
| Size: between 1-99 beds vs between 100-399 beds | 1.23 | 0.66-2.28 | 0.51 |
| Teaching hospital: major vs minor | 0.93 | 0.58-1.48 | 0.74 |
| Teaching hospital: non-teaching vs minor | 0.94 | 0.57-1.56 | 0.80 |
| Region: Midwest vs West | 0.79 | 0.47-1.33 | 0.38 |
| Region: Northeast vs West | 1.12 | 0.67-1.86 | 0.68 |
| Region: South vs West | 0.90 | 0.57-1.43 | 0.67 |
| Age | 1.00 | 0.98-1.03 | 0.74 |
| Charlson comorbidity index | 1.04 | 0.94-1.16 | 0.51 |
Revision for Periprosthetic Fracture
Hemiarthroplasty
After controlling for relevant confounding variables such as stem fixation, age, and Charlson comorbidity index, we found the risk of any revision for periprosthetic fracture after hemiarthroplasty for femoral neck fracture was higher among patients treated with cementless implants compared with those treated with cemented implants (OR 6.19 [95% CI 3.08 to 12.42]; p < 0.001). A higher Charlson comorbidity index (OR 1.16 [95% CI 1.06 to 1.28]; p = 0.002) was also a risk factor for any revision due to periprosthetic fracture. Risk factors for early revision due to periprosthetic fracture were cementless fixation (OR 7.38 [95% CI 3.17 to 17.17]; p < 0.001), major teaching hospital (OR 2.10 [95% CI 1.08 to 4.10]; p = 0.03), and a higher Charlson comorbidity index (OR 1.20 [95% CI 1.09 to 1.33]; p < 0.001) (Table 5).
Table 5.
Hemiarthoplasty model for revision for fracture
| Periprosthetic fracture | |||
| Effect | Odds ratio | 95% Wald confidence limits | p value |
| Stem fixation: cementless vs cemented | 6.19 | 3.08-12.42 | < 0.001 |
| Gender: women vs men | 0.98 | 0.63-1.53 | 0.92 |
| Size: ≥ 400 beds vs between 100-399 beds | 0.59 | 0.35-0.99 | 0.046 |
| Size: between 1-99 beds vs between 100-399 beds | 0.48 | 0.19-1.22 | 0.12 |
| Teaching hospital: major vs minor | 1.75 | 0.96-3.18 | 0.06 |
| Teaching hospital: non-teaching vs minor | 0.89 | 0.52-1.51 | 0.66 |
| Region: Midwest vs West | 0.68 | 0.36-1.31 | 0.24 |
| Region: Northeast vs West | 1.23 | 0.64-2.38 | 0.53 |
| Region: South vs West | 0.92 | 0.53-1.60 | 0.77 |
| Age | 0.99 | 0.96-1.01 | 0.35 |
| Charlson comorbidity index | 1.16 | 1.06-1.28 | 0.002 |
| Early periprosthetic fracture | |||
| Effect | Odds ratio | 95% Wald confidence limits | p value |
| Stem fixation: cementless vs cemented | 7.38 | 3.17-17.17 | < 0.001 |
| Gender: women vs men | 1.14 | 0.69-1.91 | 0.60 |
| Size: ≥ 400 beds vs between 100-399 beds | 0.50 | 0.27-0.90 | 0.02 |
| Size: between 1-99 beds vs between 100-399 beds | 0.48 | 0.17-1.35 | 0.16 |
| Teaching hospital: major vs minor | 2.10 | 1.08-4.10 | 0.03 |
| Teaching hospital: non-teaching vs minor | 0.89 | 0.50-1.61 | 0.70 |
| Region: Midwest vs West | 0.64 | 0.30-1.34 | 0.23 |
| Region: Northeast vs West | 1.17 | 0.56-2.48 | 0.67 |
| Region: South vs West | 0.96 | 0.52-1.77 | 0.89 |
| Age | 0.98 | 0.96-1.01 | 0.23 |
| Charlson comorbidity index | 1.20 | 1.09-1.33 | < 0.001 |
THA
After controlling for confounding variables, the risk of any revision for periprosthetic fracture after THA for femoral neck fracture was not higher among patients treated with cementless implants compared with those treated with cemented implants (OR 6.18 [95% CI 0.82 to 46.50]; p = 0.05). Stem fixation was not a risk factor for early revision due to periprosthetic fracture (OR 3.77 [95% CI 0.49 to 29.14]; p = 0.16) (Table 6).
Table 6.
THA model for revision for fracture
| Periprosthetic fracture | |||
| Effect | Odds ratio | 95% Wald confidence limits | p value |
| Stem fixation: cementless vs cemented | 6.18 | 0.82-46.50 | 0.05 |
| Gender: women vs men | 1.00 | 0.43-2.34 | 0.99 |
| Size: ≥ 400 beds vs between 100-399 beds | 1.66 | 0.60-4.63 | 0.33 |
| Size: between 1-99 beds vs between 100-399 beds | 2.80 | 0.77-10.22 | 0.12 |
| Teaching hospital: major vs minor | 2.13 | 0.79-5.76 | 0.14 |
| Teaching hospital: non-teaching vs minor | 0.74 | 0.21-2.60 | 0.64 |
| Region: Midwest vs West | 0.41 | 0.11-1.59 | 0.20 |
| Region: Northeast vs West | 0.80 | 0.25-2.58 | 0.71 |
| Region: South vs West | 0.71 | 0.25-2.02 | 0.53 |
| Age | 1.03 | 0.98-1.08 | 0.24 |
| Charlson comorbidity index | 1.01 | 0.77-1.31 | 0.97 |
| Early periprosthetic fracture | |||
| Effect | Odds ratio | 95% Wald confidence limits | p value |
| Stem fixation: cementless vs cemented | 3.77 | 0.49-29.14 | 0.16 |
| Gender: women vs men | 1.06 | 0.37-3.03 | 0.91 |
| Size: ≥ 400 beds vs between 100-399 beds | 1.13 | 0.35-3.69 | 0.84 |
| Size: between 1-99 beds vs between 100-399 beds | 2.57 | 0.60-11.03 | 0.20 |
| Teaching hospital: major vs minor | 1.39 | 0.39-5.04 | 0.61 |
| Teaching hospital: non-teaching vs minor | 0.65 | 0.16-2.73 | 0.56 |
| Region: Midwest vs West | 0.49 | 0.09-2.57 | 0.40 |
| Region: Northeast vs West | 1.20 | 0.29-5.04 | 0.80 |
| Region: South vs West | 1.08 | 0.31-3.74 | 0.90 |
| Age | 1.01 | 0.95-1.07 | 0.72 |
| Charlson comorbidity index | 0.82 | 0.53-1.29 | 0.39 |
Discussion
It is known that cemented stem fixation for the treatment of patients with displaced femoral neck fractures yields fewer revisions [9, 16, 18]. Avoiding revision surgery in this generally frail patient population is particularly important given the high morbidity and mortality associated with this fracture type. Despite these data, surgeons in the United States, as shown by the findings in the current study, clearly prefer cementless stem fixation. Using data from the AJRR, a nationally representative sample, we found that cementless fixation in hemiarthroplasty is a risk factor for all-cause and early revision as well as any and early revision for periprosthetic fracture. Interestingly, fixation method was not a risk factor for revision in patients treated with THA.
Limitations
Several important limitations warrant discussion. First, as we did not report data on death, deaths from “bone cement implantation syndrome” may have yielded a lower than actual risk of revision for patients treated with cemented fixation. However, given how rare death due to cement implantation syndrome is, it is unlikely to have influenced the results [11]. Second, this retrospective dataset was generated from claims data, which are susceptible to coding inaccuracies [7]. Third, the treatment groups were not balanced. Fourth, these data pertain to patients with Medicare insurance only; outcomes may be different for patients with private insurance or those living in other regions of the world. Fifth, we did not include known risk factors for revision such as femoral head size in the analysis. Sixth, the overall small number of revisions may not allow for adequate power for some comparisons. Seventh, longer follow-up may have yielded different results. Lastly, our results may have been different had we included patient-reported outcomes, as revision is not a surrogate for hip pain.
All-cause Revision
Even after controlling for relevant confounding variables such as fixation type, age, and comorbidities, we found that cementless fixation was associated with greater all-cause revision risk after hemiarthroplasty but not after THA for femoral neck fracture. The same held true for early revision. The reasons for these findings are unclear. Poor bone quality, general health status, and surgeon expertise are likely contributors. Given the finite supply of surgeons who subspecialize in hip arthroplasty surgery, it is likely that most of these operations in the United States will continue to be performed by surgeons who do not have special expertise in THA. These surgeons should consider using cemented fixation rather than cementless fixation when treating these patients with hemiarthroplasty.
Decreased age, being operated on at a major teaching hospital, and increased comorbidities were also risk factors in the hemiarthroplasty group for any and early revision. It is likely that the increased revision risk in younger patients is due to pain from acetabular chondrolysis associated with activity levels. The increased risk seen at major teaching hospitals likely stems from “sicker” patient populations seen at referral centers. Lastly, increased comorbidities may have conferred an increased risk of revision due to likely increases in factors such as infection rates and surgical site bleeding.
Revision for Periprosthetic Fracture
Revision for periprosthetic fracture was considerably more common among patients treated with cementless fixation of their hemiarthroplasty but not for those treated with THA. Early revision for periprosthetic fracture was associated with cementless fixation in hemiarthroplasty as well but not after THA. Poor bone quality, general health status, and surgeon expertise may have contributed to these findings. Hopefully, these data, representative of a national United States sample, will compel more US surgeons to use cemented fixation when performing hemiarthroplasty for the treatment of displaced femoral neck fracture.
Conclusion
This study shows that cemented fixation in hemiarthroplasty, but not in THA, for patients with displaced femoral neck fractures is associated with lower rates of all-cause revision, early revision, any revision for periprosthetic fracture, and early revision for periprosthetic fracture. Given that these data are from a nationally representative sample, more surgeons in the United States should use cemented fixation when treating these patients with hemiarthroplasty.
Footnotes
One of the authors (JWB) certifies receipt of personal payments or benefits, during the study period, in an amount of USD 100,001 to USD 1,000,000 from Zimmer Biomet; in an amount of less than USD 10,000 from Kyocera; in an amount of USD 10,000 to USD 100,000 from Flexion Therapeutics; and in an amount of USD 10,000 to USD 100,000 from Smith and Nephew.
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.
Clinical Orthopaedics and Related Research® neither advocates nor endorses the use of any treatment, drug, or device. Readers are encouraged to always seek additional information, including FDA approval status, of any drug or device before clinical use.
Ethical approval was not sought for the present study.
This work was performed at the Stanford University Medical Center, Palo Alto, CA, USA. The statistical analysis was performed at the American Academy of Orthopaedic Surgeons, Rosemont, IL, USA.
Contributor Information
Ayushmita De, Email: de@aaos.org.
Heena Jaffri, Email: jaffri@aaos.org.
John W. Barrington, Email: jbarrington@sbcglobal.net.
Paul J. Duwelius, Email: pduwelius@gmail.com.
Bryan D. Springer, Email: bryan.springer@orthocarolina.com.
References
- 1.American Hospital Association Annual Survey Database, 2015. Available at: https://www.ahadata.com Accessed September 30, 2017.
- 2.American Joint Replacement Registry. 2020 annual report. Available at: https://www.aaos.org/registries/publications/ajrr-annual-report/. Accessed May 31, 2021.
- 3.Bhandari M, Devereaux PJ, Einhorn TA, et al. Hip fracture evaluation with alternatives of total hip arthroplasty versus hemiarthroplasty (HEALTH): protocol for a multicentre randomised trial. BMJ Open. 2015;5:e006263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bhandari M, Devereaux PJ, Tornetta P, III, et al. Operative management of displaced femoral neck fractures in elderly patients: an international survey. J Bone Joint Surg Am. 2005;87:2122-2130. [DOI] [PubMed] [Google Scholar]
- 5.Bhandari M, Einhorn TA, Guyatt G, et al. Total hip arthroplasty or hemiarthroplasty for hip fracture: the HEALTH Trial. N Engl J Med. 2019;381:2199-2208. [DOI] [PubMed] [Google Scholar]
- 6.Bhandari M, Swiontkowski M. Management of acute hip fracture. N Engl J Med. 2017;377:2053-2062. [DOI] [PubMed] [Google Scholar]
- 7.Blythe R, O'Gorman PM, Crawford RW, et al. Fixation method for hip arthroplasty stem following hip fracture: a population-level cost-effectiveness analysis. J Arthroplasty. 2020;35:1614-1621. [DOI] [PubMed] [Google Scholar]
- 8.Bozic KJ, Bashyal RK, Anthony SG, Chiu V, Shulman B, Rubash HE. Is administratively coded comorbidity and complication data in total joint arthroplasty valid? Clin Orthop Relat Res.2013;471:201-205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Burgers PTPW, Van Geene AR, Van den Bekerom MPJ, et al. Total hip arthroplasty versus hemiarthroplasty for displaced femoral neck fractures in the healthy elderly: a meta-analysis and systematic review of randomized trials. Int Orthop. 2012;36:1549-1560. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hines CB. Understanding bone cement implantation syndrome. AANA J. 2018;86:433-441. [PubMed] [Google Scholar]
- 11.Hopley C, Stengel D, Ekkernkamp A, Wich M. Primary total hip arthroplasty versus hemiarthroplasty for displaced intracapsular hip fractures in older patients: systematic review. BMJ. 2010;340:c2332-c2332. [DOI] [PubMed] [Google Scholar]
- 12.Johnell O, Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int. 2006;17:1726-1733. [DOI] [PubMed] [Google Scholar]
- 13.Kristensen TB, Dybvik E, Kristoffersen M, et al. Cemented or uncemented hemiarthroplasty for femoral neck fracture? Data from the Norwegian Hip Fracture Register. Clin Orthop Relat Res. 2020;478:90-100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lewis DP, Wæver D, Thorninger R, Donnelly WJ. Hemiarthroplasty vs. total hip arthroplasty for the management of displaced neck of femur fractures: a systematic review and meta-analysis. J Arthroplasty. 2019;34:1837-1843. [DOI] [PubMed] [Google Scholar]
- 15.Liu Y, Chen X, Zhang P, Jiang B. Comparing total hip arthroplasty and hemiarthroplasty for the treatment of displaced femoral neck fracture in the active elderly over 75 years old: a systematic review and meta-analysis of randomized control trials. J Orthop Surg Res. 2020;15:215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Moerman S, Mathijssen NMC, Tuinebreijer WE, Vochteloo AJH, Nelissen RGHH. Hemiarthroplasty and total hip arthroplasty in 30,830 patients with hip fractures: data from the Dutch Arthroplasty Register on revision and risk factors for revision. Acta Orthop. 2018;89:509-514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Porter KR, Illgen RL, Springer BD, et al. Is American Joint Replacement Registry (AJRR) data representative of national data? A comparative analysis. J AAOS. 2021; in press. [DOI] [PubMed] [Google Scholar]
- 18.Rogmark C, Leonardsson O. Hip arthroplasty for the treatment of displaced fractures of the femoral neck in elderly patients. Bone Joint J. 2016;98:291-297. [DOI] [PubMed] [Google Scholar]