Femoral neck fractures in elderly patients can be devastating injuries that require medical and surgical treatment and consume considerable healthcare resources. The goal of treatment of these fractures is restoration of prefracture function without associated morbidity. Satisfactory recovery of prefracture ambulatory status correlates with younger age, comorbid medical conditions, competent mental status, male gender, community support structure, and prefracture ambulatory status. 28 When patients with femoral neck fracture are grouped without regard to treatment option, 50% to 60% regain prefracture ambulatory status. Of patients with femoral neck fractures, 10% to 20% change status from ambulatory to nonambulatory during the first year after injury. 10 This decrease in independence and ambulatory capacity is devastating to patients on a personal level, and it is expensive for the healthcare system on a socioeconomic level.
More than 125,000 femoral neck fractures occur in patients in the United States each year, and the prevalence of these fractures is expected to double by the year 2050. 10 The incidence of femoral neck fractures in patients in the United States is rising as the population ages, and it is projected that one in three persons older than 80 years will incur a fracture about the hip. 21 Femoral neck fractures account for ⅓ of hip fractures globally and approximately 50% of hip fractures in the United States. 10,21 Of patients who experience a femoral neck fracture, 50% will survive 5 years. 16 The age adjusted incidence of femoral neck fractures in the United States is 63.3 per 100,000 person-years for women and 27.7 per 100,000 person-years for men. 10
The optimal treatment for displaced femoral neck fracture in elderly patients is a matter of controversy. In this era of limited healthcare resources, it would be useful to have an algorithm for the management of femoral neck fractures in the elderly. This study was designed to evaluate the treatment of displaced femoral neck fractures (Garden Grades III and IV) in elderly patients (≥ 65 years) from the perspective of patient outcome and cost effectiveness. 5 Four surgical treatment options were analyzed: reduction with internal fixation, unipolar hemiarthroplasty, bipolar hemiarthroplasty, and total hip arthroplasty. An algorithm for treatment of these fractures is suggested using literature derived, widely applicable outcome data, and cost data from the authors’ institution.
MATERIALS AND METHODS
Outcome of fracture treatment, rates of reoperation, complications, and mortality associated with the four surgical treatments of femoral neck fracture studied were obtained from the orthopaedic literature. A principal source of information was an extensive metaanalysis performed by Lu-Yao et al, 13 which reviewed the outcomes of treatment of femoral neck fractures in 106 published reports. In addition, reviews by Koval and Zuckerman 10 and Swiontkowski 25 provide supplemental clinical insight and represent current consensus in the orthopaedic literature. These reports are not all randomized, controlled studies. However, they represent the best available data to use in a generalized application for cost effectiveness.
The current cost-effectiveness study evaluated the outcome of the four surgical treatments during a 2-year followup period. Good data exist within the first 2 years after surgery for all four surgical treatment options, and most reoperations within the internal fixation treatment group (35%) are performed within the first 2 years after surgery. 13,24 Arthroplasty failure and complication rates also are well documented within this period. In addition, the current study deals with an elderly population (≥ 65 years), with a mortality rate of 50% within 5 years of the index operation, and it is difficult to do long term cost projections on this patient population. 16
After reduction of a displaced femoral neck fracture, internal fixation is commonly accomplished with several cannulated screws. Reports on results after reduction and internal fixation of femoral neck fractures document that union is achieved in 67% of patients within 2 years, and normal blood flow to the femoral head is maintained in 85% of patients. No subsequent surgical procedures were performed in 65% of these patients. Of patients treated with reduction and internal fixation, 35% required secondary procedures: conversion to a total hip arthroplasty accounted for ⅔ removal of the internal fixation device accounted for ⅙, and repeat internal fixation accounted for ⅙. 13
Of patients treated with internal fixation for displaced femoral neck fracture, 70% achieve a pain free union during the first 2 years after injury, and 70% of the patients regain community ambulatory status after 1 year 13 (Table 1). Functional mobility and independent living potential most closely approximate prefracture status for patients successfully treated with internal fixation compared with other surgical treatments. However, this finding is valid only when patients were preselected for optimal prefracture ambulation status, independence, medical fitness, mental alertness, and proximal femoral bone stock. 20 When the Nottingham Health Profile was used, patients treated successfully with internal fixation reported fewer problems with sleep, housekeeping, hobbies, and ambulation than did patients requiring secondary total hip arthroplasty. 14 Internal fixation of appropriate femoral neck fractures provides an excellent functional outcome, and it is the standard against which other treatment options are measured.
Patients treated with unipolar hemiarthroplasty have had 80% implant survivorship 7 years after surgery. After unipolar hemiarthroplasty, 70% of patients regain community ambulatory status at 1 year, and 80% of patients report no pain or mild pain at 1 year 27 (Table 1). Unipolar hemiarthroplasty has not been successful in active elderly patients. The articulation of a large metal head on articular cartilage is associated with acetabular erosion and groin pain in active patients. Undersizing of the unipolar head is associated with increased acetabular erosion and pain. Oversized heads are associated with decreased motion. Active elderly patients with high demand requirements are not completely satisfied with these suboptimal outcomes. 24
Patients treated with bipolar hemiarthroplasty reported an 85% incidence of a pain free hip 2 years after surgery. At 2 years, 85% of patients were able to walk without aids or with only one cane. 13 Patients are permitted to bear full weight immediately after implantation, thus maximizing mobilization and the rehabilitation potential of the patient. Bipolar hemiarthroplasty is attractive in patients with neurologic impairment and instability risks, such as patients who have had a stroke and patients with Parkinson’s disease or dementia (Table 1).
Total Hip Arthroplasty
Total hip arthroplasty is one of the most enduring and successful orthopaedic interventions developed for patients with arthritis. The cost effectiveness, functional outcome, and durability of total hip arthroplasty have been documented in the orthopaedic literature. 11 The use of total hip arthroplasty for displaced femoral neck fractures traditionally has been reserved for patients with preexisting arthritic hip disease and hip pain. In addition, medical conditions that affect potential fracture union and contribute to acetabular pathology (Paget’s disease, renal osteodystrophy, osteoporosis) can be considered contraindications to hemiarthroplasty; thus, total hip arthroplasty becomes an attractive alternative.
Functionally, after femoral neck fracture, 90% of patients who underwent total hip arthroplasty reported no or slight pain 2 years after surgery. The ability to walk without aids or with only one cane is achieved by 87% of patients at 2 years or more after surgery. Survivorship of the implant was 96% within 4 years. 13 Several studies 1,7,23 have shown improvement in function, pain, and mobility when results after total hip arthroplasty were compared with results after internal fixation or hemiarthroplasty in short term followup (Table 1).
Complications and Reoperations
Perioperative complications in elderly patients treated for femoral neck fracture are common because of the high incidence of comorbidities in these patients. Increased age, dementia, malignancy, and cardiopulmonary disease have been associated with an increased risk of femoral neck fracture and an increased risk of mortality and poor outcome after surgical treatment of these fractures. 24 Urgent reduction of femoral neck fractures has been shown to reduce the risk of avascular necrosis in patients undergoing internal fixation of displaced femoral neck fractures. 24 However, mortality and morbidity rates have been shown to be higher when the medical condition of these patients is not optimized before surgery. 24
The incidence of pulmonary embolism and deep vein thrombosis is not significantly different among patients treated with internal fixation, hemiarthroplasty, or total hip arthroplasty. The rate of deep infection was found to be significantly greater for patients undergoing unipolar hemiarthroplasty. However, reports on this group of patients had a high percentage (86%) of implants performed without prophylactic antibiotics. When prophylactic antibiotics are used in a clean modern operating room, significant differences in the infection rate are not expected among the four different surgical treatment options. However, infection rates have been higher among patients undergoing arthroplasty using the posterior approach compared with other approaches to the hip for treatment of displaced femoral neck fractures. 13
Reoperation rates after surgical treatment of femoral neck fractures vary considerably among the four interventions presented. Internal fixation is associated with a median incidence of reoperation of 35% at the 2-year followup. Among these reoperations, ⅔ represent conversion to total hip arthroplasty (23%), whereas the remaining ⅓ is divided equally among repeat internal fixation (5.8%) and removal of hardware (5.8%). 13 The results of conversion total hip arthroplasty after failed primary internal fixation are better in terms of functional capacity and relief of pain than are results obtained with primary hemiarthroplasty, and they are equal to those obtained with primary total hip arthroplasty. 6,25
Treatment of these fractures by arthroplasty is associated with a risk of prosthetic dislocation. Posterolateral approaches have been associated with higher rates of mortality and may be associated with higher rates of dislocation after femoral neck fracture. 13 A metaanalysis of the orthopaedic literature revealed median dislocation rates of 2.1% for unipolar hemiarthroplasty, 2.9% for bipolar hemiarthroplasty, and 10.7% for total hip arthroplasty. When patients with one time dislocation in the acute setting (less than 4 months) are removed from the analysis, the median dislocation rate for total hip arthroplasty decreased to 2.5%. It has been recommended that an orthosis be used to limit hip range of motion in the early postoperative period to prevent acute dislocation in primary total hip arthroplasty after femoral neck fracture. 18 Dislocation after unipolar hemiarthroplasty usually is reduced with a closed procedure. However, bipolar hemiarthroplasty can be more difficult to reduce without open operative intervention.
Mortality rates for patients with displaced femoral neck fractures are similar for these four treatment methods except within the first month. The mortality associated with arthroplasty (10%) within the first 30 days after surgery is slightly higher than that associated with internal fixation (8%). However, the difference is not statistically significant. 13,24 After 1 year, mortality rates equalized among these four treatment groups. 13,24
There is a dearth of well developed cost information in the orthopaedic literature regarding the treatment of fractures of the femoral neck. In Sweden, Holmberg and Thorngren 8 looked at the consumption of hospital resources for femoral neck fractures treated with reduction and internal fixation in patients older than 50 years. The total cost of treating femoral neck fracture depended on three factors: primary operative technique, method used for rehabilitation, and the skill with which fracture complications were handled. 8 Healing complications occurred in 30% of patients. Fractures treated without complications consumed, on average, 31 cost standardized bed days, whereas fractures treated with complications consumed 101 bed days.
Parker et al 19 compared the cost benefit in quality adjusted life years of operative treatment of femoral neck fractures with nonoperative treatment of femoral neck fractures. Operative treatment proved more cost effective than nonoperative treatment. Quality adjusted life years combine the increase in survival and the improvement in health related quality of life associated with an intervention. 11 No attempt was made to compare different treatment methods for specific fracture types. Cost was calculated by adding the components of treatment, such as the operation, rehabilitation, followup clinic visits, and additional admissions, including revision surgery on the hip within 1 year of injury. 19
Laupacis et al 12 performed a prospective cost effectiveness analysis of cemented versus noncemented total hip arthroplasty performed for osteoarthritis. Costs were assessed during the first postoperative year. Costs were determined using a fully allocated costing model. Outpatient resource use was determined using patient diaries, and appropriate costs were allocated for outpatient visits, admissions to hospitals, and patient borne costs. 12
Saleh et al 22 performed an economic evaluation using a decision tree that modeled the costs and consequences of surgery versus waiting for a group of patients in need of revision hip arthroplasty. This study used a 2-year period and applied outcome probabilities from the literature to each treatment group in theoretical cost scenarios. Expenditures were derived from the literature based on preoperative and postoperative behavior for each scenario. Sensitivity analysis indicated that the conclusion is valid over a wide range of expenditures and probabilities.
The current study used a method similar to that of Laupacis et al 12 and Saleh et al 22 to compare the cost effectiveness of four surgical treatment methods for displaced femoral neck fractures. Initial hospital costs, rehabilitation costs, and costs of reoperations and complications were considered. The hospital financial records were analyzed of 185 patients with femoral neck fractures treated at the authors’ institution between October 1993 and September 1996. A resource based hospital accounting system (Decision Support System, TSI, Inc, Cambridge, MA) provided actual hospital cost data (not charges) for each procedure. The hospital costs for surgical procedures are allocated to hospital service centers, such as operating room, nursing cost, radiology, where specific, actual cost data are provided for each patient. Professional costs were not evaluated.
After being released from the hospital, 162 of 185 patients were transferred to an inpatient rehabilitation facility. Rehabilitation hospital costs were obtained from those institutions. The theoretical 2-year costs of reoperations and complications of the studied population were determined by using rates of reoperation and complications reported in the literature 1,11,13,15,24,25 applied to the cost of treating those complications at the authors’ institution (Tables 1 and 2).
After hospital costs were recorded from the resource based accounting program by the finance department’s director of decision support services (CAL), all hospital records were checked with the recovered data to ensure reliability of the data (DA, DWL). All patients whose total costs exceeded three standard deviations from the mean were excluded from the study to minimize any selection bias. A two-way sensitivity analysis was performed to calculate the cost of enabling one patient to ambulate while varying the cost and effectiveness of each intervention by 10% in a positive and negative manner (KJS). This calculation is based on the cost of enabling one patient to become ambulatory while factoring in all patient complications and costs during a 2-year followup period after intervention.
Hospital costs were determined by evaluating 185 patients, all of whom were 65 years old or older. The four surgical treatment groups included 123 (66%) patients treated with reduction and internal fixation, 16 (9%) patients treated with unipolar hemiarthroplasty, 31 (17%) patients treated with bipolar hemiarthroplasty, and 15 (8%) patients treated with total hip arthroplasty (10 hybrid total hip arthroplasty and five cemented total hip arthroplasty). The breakdown of treatment groups by years of study follows: reduction with internal fixation (34 in 1994, 42 in 1995, 47 in 1996), unipolar hemiarthroplasty (two in 1994, four in 1995, 10 in 1996), bipolar hemiarthroplasty (18 in 1994, eight in 1995, five in 1996), and total hip arthroplasty (two in 1994, four in 1995, nine in 1996). Treatment options were characterized by average length of stay, average hospital costs, operating room supplies and implant costs, and rehabilitation hospital costs (Tables 3 and 4). Rehabilitation hospital costs were determined by evaluating 162 of 185 patients who were discharged from an acute treating hospital (the authors’ institution) to a rehabilitation hospital (Table 4). Costs borne by the patients and their families not related to hospital costs were not included in this study.
Economic appraisal of healthcare programs has become increasingly prevalent in the medical literature as governments and medical policy decision makers try to allocate scarce healthcare resources. The demand for cost effectiveness and efficiency data has been increasing. Application has been far ranging, from the subsidization of pharmaceuticals to the determination of levels of Medicare reimbursement. 17 Recent literature trends have seen an increasing union of economic analysis and clinical trials. A common form of economic appraisal that has gained wide acceptance is the combination of retrospective, clinical evidence and data, with cost data from secondary sources. 4
The methodology used in the current study is of a higher quality. The cost-effectiveness data in this study are a result of a metaanalysis (rather than retrospective data from one institution). The economic costs have been accrued from one center that collected hospital cost data and resource utilization, not charges. There are two weaknesses of this analysis: (1) the reliability of the metaanalysis of Lu-Yao et al, 13 which is not based solely on prospective, randomized data (but are the most reliable data in the orthopaedic literature); and (2) the adequacy of the economic sample size and cost variability from year to year. Drummond and Davies 3 previously stated that the sample size needed to reach economic significance remains unclear. Economic effectiveness studies should collect resource information prospectively in cases in which retrospective data retrieval would be cost prohibitive or difficult to retrieve given the data collection system. It has been suggested that hospitals and systems with resource based information systems (such as the one used in this study) are the best means for achieving economic evaluations and cost-effectiveness studies. 22 Sensitivity analysis permits values to be derived with the accuracy of confidence intervals in clinical studies. If economic cost variation is seen among treatment groups to a large enough extent, significant conclusions can be reached concerning relative values through a sensitivity analysis. If economic variation is large enough, sample size differential is controlled for by the sensitivity analysis.
Assumptions and Analytical Techniques
The cost-effectiveness analysis includes actual hospital cost data for the index operation plus inpatient rehabilitation hospital costs of the patients who were not discharged home or to a nursing home. The average rehabilitation hospital cost for each treatment group was reduced by a percentage equal to the number of patients discharged home or to a nursing home. Selection bias was controlled by eliminating any patient whose total cost exceeded three standard deviations from the mean (three patients—two with reductions with internal fixation and one with unipolar hemiarthroplasty).
Several assumptions were made for this costeffectiveness analysis (Table 5). Surgical intervention is necessary for most displaced femoral neck fractures. The four surgical options include reduction with internal fixation, unipolar or bipolar hemiarthroplasty, and total hip arthroplasty. The premedical condition of all patients is the same for all four interventions. The rehabilitation protocol has differed between arthroplasty and internal fixation treatment groups. Mortality rates associated with the four interventions were not significantly different. Thus, mortality rates are not included in the analysis. 13,24 Medical morbidity is similar for all four treatment groups. Infection, deep vein thrombosis, and pulmonary embolism are associated with similar incidences among the four treatment options studied, so they were not included in the analysis. 13,24
Home physical therapy, visiting nurse care, and outpatient physician visits also were omitted from the analysis. The four operative interventions analyzed have similar physical therapy requirements. However, patients generally begin full weightbearing sooner after hip arthroplasty than patients do who have had internal fixation of femoral neck fractures. This study was not able to determine retrospectively weightbearing time tables for each patient. The omission of physical therapy information is presumed to favor internal fixation on a cost basis. A report by Koval et al 9 suggested that patients undergoing internal fixation may be able to bear weight immediately without affecting the ultimate result. Patients undergoing arthroplasty may achieve earlier functional independence than do patients undergoing internal fixation. Less physical therapy generally is needed and less cost incurred than with internal fixation.
Reoperation rates for the treatment groups were obtained from the analysis by Lu-Yao et al. 13 Median reoperation rates were adjusted for a cumulative 2-year incidence and multiplied by actual hospital cost data obtained from the authors’ institution from October 1993 to September 1996. Actual average hospital costs were determined for closed reduction of arthroplasty dislocation, removal of hip hardware, conversion total hip arthroplasty, and revision total hip arthroplasty (Table 2).
A summation of the hospital and rehabilitation costs plus the probability adjusted cost of the reoperations produced a 2-year potential cost figure that can be compared for cost-effectiveness purposes. Surgical decision making cannot completely depend on cost-effectiveness analysis for guidelines of care. Functional outcome, patient satisfaction, and patient functional requirements must influence selection of appropriate treatment for displaced femoral neck fractures.
The treatment groups were similar for all preoperative parameters. There was no statistical difference (p = 0.145) in preoperative living situation (home, independent; home, dependent; or nursing home). The number of comorbid medical conditions (p = 0.715) and mean age at the time of surgery (p = 0.1069) were not statistically different among treatment groups (Table 6).
The average hospital length of stay, hospital cost, operating room supply, and implant costs, the percentage of patients discharged to a rehabilitation hospital, average rehabilitation hospital length of stay, and average rehabilitation hospital cost were calculated for each treatment group.
From October 1993 to September 1996, 123 patients having internal fixation of femoral neck fractures using three or four 7.0-mm cannulated screws (Synthes USA, Paoli, PA) had an average length of hospital stay of 6.4 days. In 1996, the average length of stay was 5 days. Average total hospital cost was $10,383 per patient. Operating room supplies and implants totaled $649 or 6% of the total hospital cost. Of 123 patients, 102 (83%) were discharged to a rehabilitation hospital. The average rehabilitation hospital stay was 20 days, for an average rehabilitation cost of $10,655. The remainder of the 21 (17%) patients were discharged to a nursing home, their previous residence, or they died.
From October 1993 to September 1996, 16 patients having unipolar hemiarthroplasty using a modular femoral head on a cemented femoral component had an average length of hospital stay of 4.9 days. In 1996, the average length of stay was 4.2 days. Average total hospital cost was $11,554 per patient. Operating room supplies and implants totaled $1645 or 14% of the total hospital cost. Of 16 patients, 15 (94%) were discharged to a rehabilitation hospital. The average rehabilitation hospital stay was 17 days, for an average rehabilitation cost of $9190. One remaining patient died before discharge from the authors’ institution.
From October 1993 to September 1996, 31 patients having bipolar hemiarthroplasty using a modular bipolar component on a cemented femoral stem had an average length of hospital stay of 6.2 days. In 1996, the average length of stay was 6 days. Average total hospital cost was $12,371 per patient. Operating room supplies and implants totaled $2039 or 16% of the total hospital cost. Of 31 patients, 30 (97%) were discharged to a rehabilitation hospital. The average rehabilitation hospital stay was 17 days, for an average rehabilitation cost of $9190. The remainder of the patients (3%) were discharged to a nursing home.
Cemented Total Hip Arthroplasty
From October 1993 to September 1996, the average length of hospital stay for patients having total hip arthroplasty after displaced femoral neck fracture was 5.5 days. In 1996, average length of stay was 5 days. Operating room supplies and implants for cemented total hip arthroplasty totaled $1792 or 17% of the total hospital cost. The average total hospital cost for cemented total hip arthroplasty using an all polyethylene cemented acetabular component and a cemented femoral component was $10,490 per patient. Five (100%) of five patients were discharged to a rehabilitation hospital. The average rehabilitation hospital stay was 19 days, for an average rehabilitation cost of $9457.
Hybrid Total Hip Arthroplasty
From October 1993 to September 1996, the average length of hospital stay for patients having total hip arthroplasty after displaced femoral neck fracture was 5.5 days. In 1996, the average length of stay was 5 days. The average total hospital cost for hybrid total hip arthroplasty was $11,886 per patient. Operating room supplies and implant costs for hybrid total hip arthroplasty totaled $2187 or 20% of the total hospital cost. Ten (100%) of 10 patients were discharged to a rehabilitation hospital. The average rehabilitation hospital stay was 19 days, for an average rehabilitation cost of $9457.
A cost-effectiveness analysis was performed incorporating the inpatient hospital costs, the rehabilitation costs, and the probabilities of reoperation prorated over the first 2 years after surgery for all four surgical interventions. All cost figures are actual costs obtained from the authors’ institution from October 1993 to September 1996 (Tables 7–11).
On a purely cost-effectiveness basis, total hip arthroplasty represents the best value for the treatment of displaced femoral neck fractures during the first 2 years after surgery. Internal fixation represents the most expensive option when all possible operative options are considered (Table 12).
With use of the literature based comparison of clinical outcomes (Table 1) and cost data from the authors’ institution, the cost and effectiveness of each treatment option were altered by 10% on either side of the mean. A two-way sensitivity analysis was performed. The cost of having one patient ambulate for the various surgical treatment options for displaced femoral neck fracture was derived for each scenario (Table 13).
In only one most conservative scenario (the cost of reduction with internal fixation is decreased by 10% and its effectiveness of treatment is increased by 10% compared with an increase in cost of 10% for bipolar and total hip arthroplasty and a decrease in effectiveness of treatment of 10%) can reduction with internal fixation be more cost effective than bipolar and total hip arthroplasty. This scenario presupposes a total hip arthroplasty success rate of 77% and a bipolar success rate of 75% in the first 2 years, which is not consistent with reports in the literature. 6,7,10,12,15,18,23,25 The two-way sensitivity analysis supports total hip arthroplasty as the most cost effective intervention for elderly patients with displaced femoral neck fractures.
The goal of surgical treatment of displaced femoral neck fractures in elderly patients is restoration of prefracture ambulatory status with no change in hip symptoms and no complications of treatment. Cost-benefit analysis of surgical treatment for displaced femoral neck fractures has shown that all surgical treatments provide patients with improved quality adjusted life years with significant cost benefit compared with nonoperative treatment. 19
Elderly patients with displaced femoral neck fractures will achieve the best functional result with a well healed femoral neck without osteonecrosis after reduction and internal fixation. Achieving this result may be difficult. There is support in the literature for urgent reduction of these fractures to minimize the risks of avascular necrosis. 7 Patients who do not achieve a good functional result after internal fixation may have a satisfactory result after conversion total hip arthroplasty. 5,6,8 Patient selection and surgical skill remain important variables in determining the success or failure of the surgical treatment options for displaced femoral neck fractures in the elderly. These variables are difficult to extract when attempting to evaluate the orthopaedic literature on this subject.
If urgent reduction of displaced femoral neck fractures cannot be achieved because of fracture complexity, poor medical condition, or poor bone quality, the best option for patients who are ambulatory is total hip arthroplasty. Total hip arthroplasty offers a more functional, cost-effective, and enduring arthroplasty result than bipolar or unipolar hemiarthroplasty. Patients who are candidates for total hip arthroplasty should be medically optimized, and total hip arthroplasty should be performed by an experienced arthroplasty surgeon if possible. The risk of dislocation after total hip arthroplasty for displaced femoral neck fractures can be minimized with an anterior approach. The authors think total hip arthroplasty performed by experienced arthroplasty surgeons also minimizes dislocation risks and maximizes patient outcome (Table 14).
Hemiarthroplasty should be reserved for patients with debilitating medical problems and little or no ambulatory capability. Bipolar hemiarthroplasty offers little advantage over unipolar hemiarthroplasty in this patient group. 2 Bipolar hemiarthroplasty should be reserved for patients with instability risks, such as neurologic impairment (Table 14).
Total hip arthroplasty is consistently less painful, more mobile, and more functional compared with hemiarthroplasty at any followup interval. Functional results of hemiarthroplasty compare unfavorably with total hip arthroplasty as a primary treatment of displaced femoral neck fractures. Revision rates for total hip arthroplasty after femoral neck fracture are better than those reported for hemiarthroplasty (1% per year for total hip arthroplasty versus 2% per year for hemiarthroplasty). 7,18,23 The dislocation rate for total hip arthroplasty performed after femoral neck fracture is 10.7%; however, when patients with early one-time dislocation are excluded from dislocation calculations, unipolar, bipolar, and total hip arthroplasty have similar long-term dislocation rates. Reoperation rates after the surgical treatment of femoral neck fractures are least for total hip arthroplasty (12% at 3 to 6 years). 13,24 Internal fixation (35% at 2 years), unipolar hemiarthroplasty (16% at 2 years), and bipolar hemiarthroplasty (12% at 2 years) have higher reoperation rates than does total hip arthroplasty. 13,24 Cemented total hip arthroplasty and hybrid total hip arthroplasty have similar cost-effectiveness equations. The only difference in the ultimate cost involves the increased cost of a cementless acetabular component over an all polyethylene cemented component.
Opportunities for additional cost control are available if preoperative hospital days can be minimized. Although it has been shown that medical optimization before surgical intervention for femoral neck fractures improves mortality, expediting this medical optimization could decrease the preoperative hospital days that such patients frequently experience. The average length of hospital stay will be reduced additionally if total hip arthroplasty (average length of stay 5.5 days) becomes a more prevalent intervention than internal fixation (average length of stay 6.4 days). Length of stay for elective total hip arthroplasty is routinely approximately 4 days at the authors’ institution. The standardization of hip fracture care may provide similar decreased utilization benefits.
Hospital supplies and implants represent the hospital cost service center, which consumes the highest percentage of overall hospital cost regardless of the surgical treatment option selected. Unipolar (29%) and bipolar (33%) hemiarthroplasty are the largest users of the surgical options analyzed. Implant matching, negotiated vendor discount, and competitive bid implant purchasing are techniques that can be used to reduce these costs. Cemented total hip arthroplasty (18%) and hybrid total hip arthroplasty (23%) rival internal fixation (18%) for lowest percentage of overall hospital cost consumed by hospital supplies and implants. The use of three, rather than four, cannulated screws has been supported in the orthopaedic literature and would decrease the cost of internal fixation. 24
Routine hospital days and intensive care unit days in combination represent the next highest percentage of overall hospital cost by service center. These service centers represent from 27% to 34% of the total hospital cost. Better optimization of medical condition before surgical intervention, communication among the multispecialty caregivers for these patients, and early focused discharge planning can reduce hospital costs. The goal is to return these patients to a prefracture ambulatory level. Cost shifting ultimately will not increase cost effectiveness as true cost savings would.
Factors that could influence the ability to generalize information derived from this study have been minimized by use of literature derived outcome data. However, hospital cost data obtained during a 3-year period and averaged for comparison purposes has inherent limitations. Hospital length of stay trends decreased dramatically in the 1990s and in this study were not proportional across all treatment options studied. The reduction with internal fixation group showed the largest decrease in length of hospital stay among the treatment groups. However, costs of implants also have decreased significantly during the 3-year period of study and would not affect the reduction with internal fixation group. In addition, regional differences in rehabilitation hospital discharge policy could significantly affect hospital cost. However, these changes affect all treatment groups equally and should not significantly affect a comparison of hospital costs. Over a given time, any analysis will have deficiencies when there are nonstandard parameters that must be averaged to effect a comparison.
Drummond et al 4 suggested that resources consumed in medical cost effectiveness studies should be measured in actual units, rather than dollar values. However, economic evaluation should be reported using the cost from the center that contributes the largest share of patients to minimize error. In a multicenter trial, these costing issues became difficult to avoid and correct. In the current study, one center is used for costing, which avoids any data contamination. Adjustment for inflation is difficult in healthcare economic analysis because healthcare costs have not followed any of the usual indices used for public and private sector calculations. Warner and Luce 26 stated that true costs should be reported in healthcare economic studies to avoid the inaccuracies of discounting models when applied to healthcare analysis.
The deficiencies in the data presented and the biases incurred from the use of the literature have been addressed through a rigorous sensitivity analysis. The results of this analysis indicate the conclusions of the study are valid over a wide range of effectiveness probabilities. Thus, these results should be generalized to many orthopaedic centers that treat displaced femoral neck fractures.
The authors thank Health South New England Rehabilitation Hospital for their assistance and cooperation with this project.
1. Bray TJ, Smith-Hoefer E, Hooper A, et al: The displaced femoral neck fracture: Internal fixation versus bipolar endoprosthesis. Results of a prospective, randomized comparison. Clin Orthop 230:127–140, 1988.
2. Cornell CN, Levine D, O’Doherty J, et al: Unipolar versus bipolar hemiarthroplasty for the treatment of femoral neck fractures in the elderly. Clin Orthop 348:67–71, 1998.
3. Drummond MF, Davies L: Economic analysis alongside clinical trials. Revisiting the methodological issues. Int J Technol Assess Health Care 7:561–573, 1991.
4. Drummond MF, Stoddart GL, Torrence GW: Methods for the Economic Evaluation of Health Care Programs. New York, Oxford University Press 1–176, 1987.
5. Frandsen PA, Anderson E, Madsen F, et al: Garden’s classification of femoral neck fractures: An assessment of intra-observer variation. J Bone Joint Surg 70B:588–590, 1988.
6. Franzén H, Nilsson LT, Strömqvist B, et al: Secondary total hip replacement after fractures of the femoral neck. J Bone Joint Surg 72B:784–787, 1990.
7. Gebhard JS, Amstutz HC, Zinar DM, et al: A comparison of total hip arthroplasty and hemiarthroplasty for treatment of acute fracture of the femoral neck. Clin Orthop 282:123–131, 1992.
8. Holmberg S, Thorngren K-G: Consumption of hospital resources for femoral neck fracture. Acta Orthop Scand 59:377–381, 1988.
9. Koval KJ, Sala DA, Kummer FJ, et al: Postoperative weight bearing after fracture of the femoral neck or an intertrochanteric fracture. J Bone Joint Surg 80A:352–356, 1998.
10. Koval KJ, Zuckerman JD: Hip fractures: 1. Overview and evaluation and treatment of femoral neck fractures. J Am Acad Orthop Surg 1:141–149, 1994.
11. Laupacis A, Bourne R, Rorabeck C, et al: The effect of elective total hip replacement on health-related quality of life. J Bone Joint Surg 75A:1619–1626, 1993.
12. Laupacis A, Bourne R, Rorabeck C, et al: Costs of elective total hip arthroplasty during the first year: Cemented versus noncemented. J Arthroplasty 9:315–487, 1994.
13. Lu-Yao GL, Keller RB, Littenberg B, et al: Outcomes after displaced fractures of the femoral neck: A meta-analysis of one hundred and six published reports. J Bone Joint Surg 76A:15–25, 1994.
14. Marcus RE, Heintz JJ, Pattee GA: Don’t throw away the Austin Moore. J Arthroplasty 7:31–36, 1992.
15. Nilsson LT, Franzén H, Strömqvist B, et al: Function of the hip after femoral neck fractures treated by fixation or secondary total hip replacement. Int Orthop (SICOT) 15:315–318, 1991.
16. Nilsson LT, Jalovaara P, Franzén H, et al: Function after primary hemiarthroplasty and secondary total hip arthroplasty in femoral neck fracture. J Arthroplasty 9:369–374, 1994.
17. O’Brien BJ, Drummond MF, Labelle RJ, et al: In search of power and significance: Issues in the design and analysis of stochastic cost-effectiveness studies in health care. Med Care 32:150–163, 1994.
18. Papandrea RF, Froimson MI: Total hip arthroplasty after acute displaced femoral neck fractures. Am J Orthop 2:85–88, 1996.
19. Parker MJ, Myles JW, Anand JK, et al: Cost-benefit analysis of hip fracture treatment. J Bone Joint Surg 74B:261–264, 1992.
20. Robinson CM, Saran D, Annan IH: Intracapsular hip fractures: Results of management adopting a treatment protocol. Clin Orthop 302:83–91, 1994.
21. Salai M, Rosenberg E: Multivariant analysis on the long-term results of hemiarthroplasty after subcapital fracture of the femur. J Am Geriatr Soc 44:477–478, 1996. Letter.
22. Saleh KJ, Wood KC, Gafni A, et al: Immediate surgery versus waiting list policy in revision total hip arthroplasty: An economic evaluation. J Arthroplasty 12:1–10, 1997.
23. Skinner P, Riley D, Ellery J, et al: Displaced subcapital fractures of the femur: A prospective randomized comparison of internal fixation, hemiarthroplasty and total hip replacement. Injury 20:291–293, 1989.
24. Swiontkowski MF: Intracapsular Hip Fractures. In Browner BD, Jupiter JB, Levine AM, Trafton PG (eds). Skeletal Trauma: Fractures, Dislocations, Ligamentous Injuries. Philadelphia, WB Saunders 1369–1442, 1992.
25. Swiontkowski MF: Current concepts review: Intracapsular fractures of the hip. J Bone Joint Surg 76A:129–138, 1994.
26. Warner KE, Luce BR: Cost Benefit and Cost Effectiveness Analysis in Health Care: Principles, Practice and Potential. Ann Arbor, Health Administration Press 1982.
27. Wathne RA, Koval KJ, Aharonoff GB, et al: Modular unipolar versus bipolar prosthesis: A prospective evaluation of functional outcome after femoral neck fracture. J Orthop Trauma 9:298–302, 1995.
© 2001 Lippincott Williams & Wilkins, Inc.
28. Young Y, German P, Brant L, et al: The predictors of surgical procedure and the effects on functional recovery in elderly with subcapital fractures. J Gerontol A Biol Sci Med Sci 51:M158–M164, 1996.