The results of the patient questionnaires completed at four, twelve, and twenty-four months are summarized in Table III (fixation compared with hemiarthroplasty), Table IV (fixation compared with total hip replacement), and Table V (hemiarthroplasty compared with total hip replacement). The fixation group had a significantly worse mean score on the hip-rating questionnaire than both arthroplasty groups at four and twelve months, and this worse overall score reflected poorer scores for all of the subscales. The observed differences in the overall scores were close to the 7 points that had been judged to be clinically important when the sample size was decided on, although it should be noted that the 95% confidence intervals indicated that the true difference could be larger or smaller. The patients allocated to be treated with total hip replacement had, on the average, a better overall score than those allocated to be treated with hemiarthroplasty, and this difference was significant at twenty-four months (adjusted difference in overall score, –6.45; 95% confidence interval, –12.53 to –0.37). The largest contributions to this difference came from the walking and function subscores. The results of the prespecified subgroup analyses suggested that all of these differences between the groups were more pronounced in patients who were between sixty and seventy-four years of age than they were in those who were seventy-five years of age or older.
The results obtained with the EuroQol general health measure followed a broadly similar pattern. At four and twelve months, those allocated to be treated with total hip replacement had the best scores, and those allocated to be treated with fixation had the worst scores. At twenty-four months, however, the bipolar hemiarthroplasty group had the lowest score, which was significantly worse than the score in the total hip replacement group.
Table VI presents unadjusted mean costs and cost differences by resource category over two years. Although fixation was initially less costly than bipolar hemiarthroplasty, this short-term cost advantage was eroded by the significantly increased costs of subsequent hip-related admissions. The cumulative additional costs over two years for all hip-related episodes following fixation was £3504 higher (95% confidence interval, £1159 to £5851) than the cumulative costs for bipolar hemiarthroplasty. No significant difference was found in either the costs of the initial inpatient episode or the costs of the non-hip-related admissions following the initial episode. There was a similar pattern of cost differences between the fixation and total hip replacement groups, although only the difference in hip-related-admission costs was significant. In the comparison between bipolar hemiarthroplasty and total hip replacement, the confidence intervals around the hip-related admission costs were wide, reflecting the small numbers of patients who were readmitted to the hospital (seven and eight, respectively), and we found no significant difference between these two groups. Additional analyses were performed that (1) adjusted for age, sex, and whether the randomization had been two or three-way and (2) varied the cost of hip-related admissions and prostheses. These adjustments had little impact on the findings.
The most striking finding of this study was the poor outcome in patients treated with reduction and fixation. This procedure was associated with a high rate of revision surgery and an inferior functional outcome compared with that of arthroplasty. This trend was particularly evident for younger patients (sixty to seventy-four years old). Although reduction and fixation had the lowest acute-admission costs (with less expensive implants, shorter operative time, and shorter initial hospital stays), the greatly increased need for readmissions and reoperations resulted in this management option having the highest costs overall. Differences between the two types of arthroplasty were less marked; however, the functional outcome at two years was significantly better following total hip replacement. Although other randomized studies of the management of displaced intracapsular hip fractures have been reported2-6,13-23, we believe that ours is the first to focus on healthy patients who were previously capable of independent mobility. We believe that we are also the first to incorporate validated functional outcome measures and an economic evaluation.
The internal validity of the study was high. Care was taken to minimize bias by using a centralized method of randomization and by excluding participating surgeons from data collection and analysis. The groups that we compared were similar at baseline, follow-up was unusually complete, the analysis was based on intention to treat, and the participants themselves assessed their outcome with use of standardized questionnaire-based instruments at up to two years postoperatively.
Although this trial was large for an orthopaedic study, the estimated differences are statistically imprecise, and this is an important limitation. It applies in particular to the comparisons involving total hip replacement because some surgeons were not willing to randomize their patients to be treated with what they believed was an inappropriately complex procedure. The sample size was based on the judgment that it was clinically important to identify differences in the hip-rating-questionnaire score of approximately 7 points. The observed differences between fixation and arthroplasty at four and twelve months, and between total hip replacement and hemiarthroplasty at twenty-four months, were close to this value, but there were wide confidence intervals, indicating that the true differences could have been larger or smaller. Imprecision was a particular problem for statistically rare clinical events, such as death.
The main issue in the interpretation of the findings in this trial is their generalizability. To an extent, this is an issue in all randomized controlled trials; patients who are eventually recruited have gone through a selection process, and they differ from those who are not recruited. It is then a matter of judgment whether those recruited are sufficiently similar to patients who are cared for by another surgeon for the results to be generalized to that practice. The three procedures compared in our trial were markedly different, and, because of these differences, surgeons often had strong views about which patients should or should not be treated with each of the three operations. This was particularly true for total hip replacement. Some surgeons indicated that they would not be willing to randomize any patient to be treated with this procedure. (They were, however, usually sufficiently uncertain regarding the comparative benefits of fixation and hemiarthroplasty to randomize patients to receive either of those procedures.) Other surgeons, while agreeing to recruit patients into the trial, sometimes had strong views on the management of individual patients so some potentially eligible patients were deemed ineligible. There are other reasons why recruitment into a trial like ours will always be difficult. These patients are often frail and in pain, and they are admitted to the hospital on an emergency basis at all times of the day. Securing informed consent for entry into a trial is therefore unusually difficult and time-consuming. This can lead to eligible patients being missed. About one in eight patients who were seen for a displaced intracapsular hip fracture at the trial centers were considered eligible for entry into the trial. All patients had to have been previously healthy and mobile, and alternative procedures had to be considered clinically appropriate for them. Recruitment was therefore slow, and the results can be generalized only to these types of patients.
Not all participants received the operation to which they had been randomly allocated. The reasons were clinical (e.g., an inability to reduce a fracture) or organizational (e.g., the institution was not equipped to perform the allocated procedure or the equipment failed). However, all analyses were based on the allocation groups (the intention-to-treat principle) to avoid biased selection of groups for comparison. Secondary observational analyses based on the actual operation performed showed a slight widening of the differences observed in the intention-to-treat analyses.
There was some variation in surgical practice due to the multicenter design of the study. However, all patients in the reduction-and-fixation group were treated with either cannulated screws or a sliding hip screw, and there was no evidence that the type of implant had a bearing on the results. Similarly, all of the arthroplasties were performed with cemented components, and the total hip implants were either Exeter or Charnley designs, both of which have been documented as having satisfactory long-term results. Analysis revealed no relationship between the result and the specific implant used, although patients who had the arthroplasty performed through a posterior exposure had a higher rate of dislocation (four of fourteen compared with one of 112).
Other recent randomized studies have also demonstrated high reoperation rates (34% to 43%) following reduction and fixation of displaced intracapsular hip fractures5,6,22,23. The most common reasons for the reoperations were fixation failure and nonunion. Although osteonecrosis is a well-recognized complication of this fracture, it is not the most common cause of reoperations. The other randomized studies included patients with limited mobility or cognitive function, and it is often assumed that healthy older patients have a lower complication rate following reduction and fixation. However, in our study of a healthy group of patients, reduction and fixation had a failure rate similar to those reported in the other studies. The poorer functional outcome was particularly marked in the younger patients. However, the timing of the follow-up could have influenced the findings with regard to the reoperations. Whereas fixation failures commonly occurred soon after the surgery, there might be disproportionately larger numbers of failures of the arthroplasties beyond the current follow-up period of two years. An additional potential confounding variable is the experience of the surgeons who performed the internal fixation procedures compared with that of the surgeons who did the arthroplasties.
The rates of mortality and general complications did not differ significantly across the study groups. Other randomized trials have revealed similar findings24. Although reduction and fixation is a shorter, less invasive surgical procedure, many patients with displaced intracapsular hip fracture have medical co-morbidities that contribute to the risk of general complications. In addition, the need for revision surgery in the fixation group may have contributed to the similarity of the mortality rates among the three groups at the two-year follow-up point. This would be consistent with the finding of a recent meta-analysis that one-year mortality rates were similar following different types of treatment of these fractures (despite a suggestion of lower rates after fixation at four months)24.
One question raised by our findings is whether the poorer functional outcome after reduction and fixation just reflects the failures that required additional surgery. At twenty-four months, the functional outcome for the patients who were managed with fixation and did not have subsequent surgery was clearly better than the outcome for those managed with fixation who did have subsequent surgery. However, even those who did not have additional surgery had a mean score that was lower than that for the total hip replacement group.
The best clinical and functional outcomes in this study were observed after total hip replacement. This has not been a popular method of treating these fractures in the past, at least in part because of a perception that it is associated with an unacceptably high rate of prosthetic dislocation. We did not find this to be the case, and the recent meta-analysis showed a mean rate of dislocation of 6.9%24. This is certainly higher than what is expected after arthroplasty for primary osteoarthritis, but we believe that it is still acceptably low.
The reason for the apparent functional deterioration in the bipolar hemiarthroplasty group in our trial is not clear. Chance cannot be ruled out without further follow-up. Alternatively, it may be explained by early acetabular erosion associated with the bipolar head. As the study design did not include long-term radiographic follow-up, any further comment on this possibility would be speculative.
In our cost estimates, we accounted for the resource consequences of the initial procedure following the injury, subsequent hip-related and non-hip-related episodes, and the use of other health services. The costs reflected contemporary techniques employed by orthopaedic surgeons and utilization of health services by patients allocated to one of three treatment groups. Although we believe that our methods for estimating the differences in health service costs are robust, the absolute and relative differences primarily reflect the numbers of hip-related readmissions. While there were marked differences between the number of readmissions following fixation and those following both types of arthroplasty, which were reflected in the significant differences in costs, there were relatively few readmissions after either bipolar or total hip arthroplasty (seven compared with eight). The observed difference in cost between the two arthroplasty groups, which was not significant, reflected the high costs of the readmissions following bipolar hemiarthroplasty. The differences also reflect observed practice in Scottish centers between 1996 and 2002 and the specific resource unit costs recorded in our study. Both resource use and cost will vary across different health-care systems and may change over time as novel techniques are adopted and new patterns of service use become established.
The poorer outcomes and higher costs following fixation indicate that it is not cost-effective compared with either bipolar hemiarthroplasty or total hip replacement; this conclusion should remain true unless there is a dramatic convergence in revision rates across the procedures.
On the basis of our findings, we no longer recommend reduction and fixation for the treatment of previously healthy older patients with a displaced intracapsular hip fracture. Arthroplasty with cement is our treatment of choice. In this study, the two-year results of total hip replacement appeared to be better than those of bipolar hemiarthroplasty, but this finding was based on relatively small numbers of patients; hence, other important differences between the treatment groups could not be ruled out. Ideally, this comparison should be replicated in other trials. Longer-term follow-up of the patients treated with arthroplasty in this study might allow identification of the reasons for the deterioration in function observed at two years following the bipolar hemiarthroplasties. It would also determine whether the satisfactory outcomes of total hip replacement are maintained.▪
A commentary is available with the electronic versions of this article, on our web site () and on our quarterly CD-ROM (call our subscription department, at 781-449-9780, to order the CD-ROM).
In support of their research for or preparation of this manuscript, the authors received a grant from the National Health Service R&D Health Technology Assessment Programme. The Health Services Research Unit is core funded by the Chief Scientist Office of the Scottish Executive Health Department. None of the authors received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.
NOTE: The authors acknowledge the participation of surgeons in the following orthopaedic units: Aberdeen Royal Infirmary, Dumfries and Galloway Royal Infirmary, Edinburgh Royal Infirmary, Falkirk Royal Infirmary, Glasgow Royal Infirmary, Western Infirmary Glasgow, Law Hospital Lanarkshire, Ninewells Hospital Dundee, Perth Royal Infirmary, Queen Margaret Hospital Dunfermline, and Royal Alexandra Hospital Paisley.
1. , McLauchlan GJ, Robinson CM, Christie J. Epidemiology of fractures in 15,000 adults: the influence of age and gender. J Bone Joint Surg Br. 1998;80: 243-8.
2. , Mölster A, Raugstad TS. Internal fixation versus primary prosthetic replacement in acute femoral neck fractures: a prospective, randomized clinical study. Br J Surg. 1979;66: 56-60.
3. , Barrington R. Internal fixation versus hemiarthroplasty for the displaced subcapital fracture of the femur. A prospective randomised study. J Bone Joint Surg Br. 1981;63: 357-61.
4. , Pryor GA. Internal fixation or arthroplasty for displaced cervical hip fractures in the elderly: a randomised controlled trial of 208 patients. Acta Orthop Scand. 2000;71: 440-6.
5. , Calder SJ, Anderson GH, Ward G, Jagger C, Harper WM, Gregg PJ. Treatment for displaced intracapsular fracture of the proximal femur. A prospective, randomised trial in patients aged 65 to 79 years. J Bone Joint Surg Br. 2001;83: 206-12.
6. , Carlsson Å, Johnell O, Sernbo I. A prospective randomised trial of internal fixation versus arthroplasty for displaced fractures of the neck of the femur. Functional outcome for 450 patients at two years. J Bone Joint Surg Br. 2002;84: 183-8.
7. . The management of intracapsular fractures of the proximal femur. J Bone Joint Surg Br. 2000;82: 937-41.
8. , Folstein SE, McHugh PR. “Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12: 189-98.
9. , McPherson GC, Ramsay CR, Campbell MK. The method of minimization for allocation to clinical trials: a review. Control Clin Trials. 2002;23: 662-74.
10. , Charlson ME, Szatrowski TP, Ranawat CS. A self-administered hip-rating questionnaire for the assessment of outcome after total hip replacement. J Bone Joint Surg Am. 1992;74: 587-97.
11. EuroQol—a new facility for the measurement of health-related quality of life. The EuroQol Group. Health Policy. 1990;16: 199-208.
12. , Zethraeus N, Svensson O, Tornkvist H, Ponzer S. Femoral neck fractures in the elderly: functional outcome and quality of life according to EuroQol. Qual Life Res. 2002;11: 473-81.
13. , Rasmussen T, Christensen S, Holm-Moller S, Lauritzen J. Internal fixation or prosthetic replacement in fresh femoral neck fractures. Acta Orthop Scand. 1984;55: 712.
14. , Riley D, Ellery J, Beaumont A, Coumine R, Shafighian B. Displaced subcapital fractures of the femur: a prospective randomized comparison of internal fixation, hemiarthroplasty and total hip replacement. Injury. 1989;20: 291-3.
15. , Oosterwijk WM, Goris RJ. Osteosynthesis versus endoprosthesis in the treatment of unstable intracapsular hip fractures in the elderly. A randomised clinical trial. Arch Orthop Trauma Surg. 1993;113: 39-45.
16. , Sernbo I, Carlsson A, Fredin H, Johnell O. Social function after cervical hip fracture. A comparison of hook-pins and total hip replacement in 47 patients. Acta Orthop Scand. 1996;67: 431-4.
17. , Adolphson P, von Sivers K, Dahlborn M, Dalen N. Bone and muscle mass after femoral neck fracture. A controlled quantitative computed tomography study of osteosynthesis versus primary total hip arthroplasty. Arch Orthop Trauma Surg. 1997;116: 470-4.
18. , Marsh G. Internal fixation versus hemiarthroplasty versus total hip arthroplasty for displaced subcapital fractures of femur—13 year results of a prospective randomised study. Injury. 2000;31: 793-7.
19. , Jacobsson SA, Ivarsson I, Knutsson A, Wahlstrom O. Internal fixation versus total hip arthroplasty in the treatment of displaced femoral neck fractures: a prospective randomized study of 100 hips. Acta Orthop Scand. 2000;71: 597-602.
20. , Douw CM, van Breukelen AM, Laurens DR, Mulder PG, Wereldsma JC, van Vugt AB. Cannulated screws versus hemiarthroplasty for displaced intracapsular femoral neck fractures in demented patients. Ann Chir Gynaecol. 2000;89: 132-7.
21. , Laine HJ, Tarvainen T, Aho H. Thompson hemiarthroplasty is superior to Ullevaal screws in treating displaced femoral neck fractures in patients over 75 years. A prospective randomized study with two-year follow-up. Ann Chir Gynaecol. 2001;90: 225-8.
22. , Khan RJ, Crawford J, Pryor GA. Hemiarthroplasty versus internal fixation for displaced intracapsular hip fractures in the elderly. A randomised trial of 455 patients. J Bone Joint Surg Br. 2002;84: 1150-5.
23. , Ponzer S, Svensson O, Soderqvist A, Tornkvist H. Internal fixation compared with total hip replacement for displaced femoral neck fractures in the elderly. A randomised, controlled trial. J Bone Joint Surg Br. 2003;85: 380-8.
Copyright 2006 by The Journal of Bone and Joint Surgery, Incorporated
24. , Devereaux PJ, Swiontkowski MF, Tornetta P 3rd, Obremskey W, Koval KJ, Nork S, Sprague S, Schemitsch EH, Guyatt GH. Internal fixation compared with arthroplasty for displaced fractures of the femoral neck. A metaanalysis. J Bone Joint Surg Am. 2003;85: 1673-81.