 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1998 The Journal of Bone and Joint Surgery, Inc. Total Hip Arthroplasty for the Treatment of an Acute Fracture of the Femoral Neck. Long-term Results* , ROCHESTER, MINNESOTAInvestigation performed at the Mayo Clinic, Rochester
We reviewed the long-term results of 126 consecutive total hip arthroplasties performed with cement in eighteen men and 108 women who had an acute fracture of the femoral neck. The patients had a mean age of seventy-five years (range, thirty-nine to eighty-nine years) at the time of the operation and were followed for a minimum of 10.1 years (or until the patient died or had a revision operation) and a maximum of 20.4 years. The median duration of follow-up was 8.8 years for all patients and 15.7 years for the twenty-two patients who were alive at the end of the study period. Six hips (5 per cent) were revised because of aseptic loosening. Survivorship analysis revealed that the probability of survival of the prosthesis without revision (with 95 per cent confidence intervals) was 95 per cent (91 to 99 per cent) at five years, 94 per cent (88 to 98 per cent) at ten years, 89 per cent (79 to 97 per cent) at fifteen years, and 84 per cent (66 to 97 per cent) at twenty years. Of the 118 patients who were alive at the one-year postoperative examination, 117 (99 per cent) had no pain or mild pain and eighty-one (69 per cent) had regained or had an improvement in the preoperative level of function. At the latest follow-up examination, eighty-seven (86 per cent) of the 102 patients who were available still had no pain or only mild pain. Twenty-six patients (21 per cent) had had perioperative medical complications, and twenty-one patients (17 per cent) had had operative complications, including thirteen patients (10 per cent) who had had a dislocation of the hip. Total hip arthroplasty performed in elderly patients for the treatment of an acute fracture of the femoral neck was associated with a higher rate of complications than usually is reported for hemiarthroplasty in such patients. However, the total hip arthroplasty provided good clinical results and was associated with long-term survival of the prosthesis.
Treatment of an acute displaced fracture of the femoral neck in elderly patients remains controversial33; some authors have advocated open reduction and internal fixation and others, prosthetic replacement of the femoral head3,16,18,28,34. When prosthetic replacement of the femoral head is chosen, hemiarthroplasty (unipolar or bipolar) rather than total hip arthroplasty is preferred in most circumstances. The advantages of hemiarthroplasty, compared with total hip arthroplasty, for most patients who have a fracture of the femoral neck include a shorter operative time (because acetabular resurfacing is not necessary) and a lower risk of early complications (most notably dislocation). Nevertheless, total hip arthroplasty may be considered for some elderly patients who have an acute fracture of the femoral neck, most often those who have preexisting symptomatic arthrosis of the hip13,22,23,25. The long-term results of total hip arthroplasty for the treatment of an acute fracture of the femoral neck have not been reported previously for a large series of elderly patients, to our knowledge. A better understanding of the long-term outcome of all of the methods available for the treatment of a displaced fracture of the femoral neck is needed to optimize the management of this common clinical problem24. To learn more about the long-term results of total hip arthroplasty performed with cement for the treatment of an acute fracture of the femoral neck, we reviewed the records of 126 consecutive patients who had had such a procedure between 1970 and 1981. The early results for these patients have been reported previously29.
One hundred and twenty-six consecutive total hip arthroplasties (126 patients) performed between 1970 and 1981 for the treatment of an acute fracture of the femoral neck were identified with use of the joint registry at our institution. We retrospectively reviewed the information in the patient's charts and radiographs. Clinical and radiographic information on all patients had been prospectively collected as a routine part of monitoring of all arthroplasties at our institution. The patients included eighteen men and 108 women who had a mean age of seventy-five years (range, thirty-nine to eighty-nine years) at the time of the index arthroplasty. One patient was less than fifty years old, eight were fifty to fifty-nine years old, thirty-one were sixty to sixty-nine years old, fifty-one were seventy to seventy-nine years old, and thirty-five were eighty to eighty-nine years old. None of the patients had a bilateral procedure. The indications for prosthetic replacement rather than internal fixation included a disease affecting the hip and a fracture with characteristics (particularly displacement with marked posterior comminution and osteopenia) that have been associated with unfavorable results of internal fixation. Eighteen patients (14 per cent) had a disease affecting the hip: nine had osteoarthrosis, five had rheumatoid arthritis, and four had Paget disease. The fractures were categorized by one of us (B. P. H. L.) according to Garden's classification system9. There were fifteen type-II, forty-one type-III, and seventy type-IV fractures. Radiographic evidence of osteopenia was graded by the same one of us, according to the index of Singh et al.30. An index of 3 points or less was considered to be osteopenic. On the basis of this grading system, eighty-two hips (65 per cent) were markedly osteopenic. All total hip arthroplasties were performed with insertion of the acetabular and femoral components with cement. Six prosthetic designs were used: Charnley components (DePuy, Warsaw, Indiana) were implanted in sixty-nine hips; Trapezoidal-28 components (Zimmer, Warsaw, Indiana), in twenty-four; Charnley-Müller components (DePuy) in seventeen; Aufranc-Turner components (Howmedica, Rutherford, New Jersey), in ten; HD-2 components (Howmedica), in three; and Protosul components (DePuy), in three. The size of the femoral head was twenty-two millimeters in sixty-nine hips, thirty-two millimeters in thirty-two, twenty-eight millimeters in twenty-four, and twenty-six millimeters in one. The femoral component was inserted with use of so-called first-generation cementing techniques in 114 patients and with use of so-called second-generation techniques in twelve. The total hip arthroplasty was performed through a transtrochanteric approach in sixty-nine hips, an anterolateral approach in fifteen, a direct lateral approach in twenty-five, and a posterior approach in seventeen. Postoperatively, the patients were evaluated at three months, one year, and two years after the operation and then on a regularly scheduled basis thereafter. Standardized operative, clinical, and radiographic data on each patient had been collected prospectively, but the data were collated retrospectively at the time of the study. The latest evaluation was performed with a physical examination for fifty-four patients (43 per cent), by means of a detailed questionnaire for sixty (48 per cent), and by telephone interview for twelve (10 per cent). Of the twenty-two patients who were alive at the end of the study period, twelve were most recently evaluated with a physical examination and an interview in the physician's office, eight returned a questionnaire sent by mail, and two were interviewed by telephone. The number of questionnaires reflected the number of patients who were unable to travel for follow-up examinations because of age and associated medical conditions. Radiolucent lines in each of the three acetabular zones described by DeLee and Charnley7 were measured on standard anteroposterior and lateral radiographs of the hip. Fixation of the acetabular component was categorized according to an adaptation of the criteria described by Hodgkinson et al.15. A non-continuous radiolucent line less than two millimeters wide in at least one zone at the bone-cement interface was considered to indicate possible loosening; a non-continuous radiolucent line at least two millimeters wide in at least one zone, probable loosening; and a continuous radiolucent line of any width, migration of the component, or fracture of the cement, definite loosening. Fixation of the femoral component was evaluated in accordance with the categories of loosening described by Harris et al.14. A radiolucent line occupying 50 to 99 per cent of the bone-cement interface was considered to indicate possible loosening; a complete radiolucent line at the bone-cement interface, probable loosening; and migration of the component or fracture of the cement mantle, definite loosening. Evidence of so-called debonding and a radiolucent line that was less than two millimeters wide between the shoulder of the prosthesis and the cement were not considered to indicate loosening.
None of the patients died during the operation. One patient died from a myocardial infarction thirty-four days after the operation. The rate of mortality was 6 per cent (eight patients) at one year, 8 per cent (ten patients) at two years, 28 per cent (thirty-five patients) at five years, 56 per cent (seventy-one patients) at ten years, and 74 per cent (ninety-three patients) at fifteen years. All patients were followed for a minimum of 10.1 years (maximum, 20.4 years) or until the patient died or had a revision operation. No patient was lost to follow-up. At the time of the latest follow-up, twenty-two patients were alive and 104 had died. The median duration of follow-up was 8.8 years (range, thirty-four days to 20.4 years) for all patients and 15.7 years (range, 10.1 to 20.4 years) for the twenty-two patients who were alive at the end of the study period. Forty-four patients (35 per cent) had a preexisting condition that limited their activity. Thirty-six of these patients walked with use of an assistive device: twenty-seven patients used a cane, six used crutches, and three used a walker. Of the 118 patients who were alive one year after the operation, ninety-eight (83 per cent) had no pain, nineteen (16 per cent) had mild pain, and one patient (1 per cent) had moderate-to-severe pain. Sixty-four patients (54 per cent) used an assistive device for walking: forty-nine used a cane, five used crutches, and ten used a walker. Eighty-one patients (69 per cent) had the same or an increased level of activity compared with the preoperative status, and thirty-seven (31 per cent) had a decreased level. The decrease in activity was attributable at least in part to deterioration associated with other medical conditions. One hundred and two patients were available for follow-up at least three years postoperatively. Sixty-seven (66 per cent) had no pain; twenty (20 per cent), mild pain; thirteen (13 per cent), moderate pain; and two (2 per cent), severe pain. Six hips were revised because of aseptic loosening. Three had revision of the femoral component only, one had revision of the acetabular component only, and two had a revision of both the femoral and the acetabular component. The rate of survival of the components without revision (with 95 per cent confidence intervals) was 95 per cent (91 to 99 per cent) at five years, 94 per cent (88 to 98 per cent) at ten years, 89 per cent (79 to 97 per cent) at fifteen years, and 84 per cent (66 to 97 per cent) at twenty years (Fig. 1). At the time of the primary total hip arthroplasty, the mean age of the patients who ultimately needed a revision was 66.7 years. Of those patients, two were between fifty and fifty-nine years old, two were between sixty and sixty-nine years old, and two were between seventy and seventy-nine years old.
The median duration of radiographic follow-up was 5.3 years for the entire group and 11.7 years for the twenty-two patients who were alive at the end of the study. On the most recent radiographs, 104 acetabular components (83 per cent) had no evidence of loosening, fourteen (11 per cent) were possibly loose, three (2 per cent) were probably loose, and five (4 per cent) were definitely loose. One hundred and nine femoral components (87 per cent) had no evidence of loosening, nine (7 per cent) were possibly loose, two (2 per cent) were probably loose, and six (5 per cent) were definitely loose. Seventy-eight patients (62 per cent) had a preoperative medical comorbidity. Twenty-six patients (21 per cent) had a perioperative medical complication, including cardiac complications (nine patients), deep venous thrombosis or pulmonary embolism (six patients), renal complications (seven patients), and gastrointestinal complications (four patients). Twenty-one patients (17 per cent) had an operative complication; the most notable was dislocation of the hip (thirteen patients). Other complications included an infected hematoma (one patient), symptomatic heterotopic ossification (two patients), and non-union of the greater trochanter (five patients). None of the patients had an infection associated with the prosthetic joint. Four patients had an additional operative procedure, including dëbridement of an infected hematoma (one patient), excision of heterotopic bone (two patients), and open reduction of a dislocation (one patient). Eleven of the thirteen dislocations were treated successfully with closed reduction. One patient had recurrent dislocation that was not treated operatively, and one patient had an open reduction four months after the total hip arthroplasty. A dislocation occurred in three of the forty hips that had had an anterolateral or direct lateral approach, in six of the sixty-nine hips that had had a transtrochanteric approach, and in four of the seventeen hips that had had a posterior approach. Although the posterior approach appeared to be associated with a higher rate of dislocation than the other approaches, the difference was not found to be significant, with the numbers available (p = 0.18). Also, with the numbers available, no significant association was identified between a small femoral head (twenty-two millimeters compared with all other sizes) and an increased risk of dislocation (p = 0.48).
The treatment of displaced fractures of the femoral neck in elderly patients has evolved, but it remains controversial1-3,16-18,28,32,34,35. Those in favor of open reduction and internal fixation have mentioned the preservation of the hip joint and the avoidance of complications related to prosthetic implantation as advantages, whereas those in favor of replacement with a prosthetic femoral head have considered its advantages to include a shortened duration of postoperative rehabilitation and the avoidance of problems related to healing of the fracture and to vascularity of the femoral head3,17,18,28,34. Prosthetic replacement for the treatment of an acute fracture of the femoral neck usually consists of hemiarthroplasty. However, in certain patients, such as those who have preexisting arthrosis of the hip or who anticipate a high level of activity (walking about the community), a total hip arthroplasty also may be considered6,8,10-12,29,34. Little information is available concerning the long-term outcome of total hip arthroplasty in such patients. Elderly patients who have a fracture of the femoral neck typically have a limited level of activity and a limited life expectancy, both of which would be expected to improve the chance of survival of the prosthesis. However, the same patients also have poorer-quality bone, which might decrease the durability of prosthetic fixation. The goal of the present study was to document the long-term results of total hip arthroplasty for the treatment of an acute fracture of the femoral neck. The strengths of the present study include the large number of patients, the long-term follow-up, and the high percentage of patients who were followed for a long period of time. The primary limitation was a function of the long-term nature of the study. As the indications for a total hip arthroplasty for treatment of a fracture of the femoral neck have changed since the patients were entered into the study, the percentage of patients in the present series who had normal hips before the fracture is higher than the percentage at our institution at present. Nevertheless, the findings of the present study can be extrapolated to the present population of patients in whom the procedure might be used. Whether or not they have a preexisting disease affecting the hip, elderly patients who have an acute fracture of the femoral neck have many similarities—that is, a limited life expectancy, a limited level of activity, and poorer-quality bone. The advantages of hemiarthroplasty, compared with total hip arthroplasty, for the treatment of a fracture of the femoral neck include the more limited nature of the procedure (because acetabular resurfacing is not necessary) and the lower risk of instability of the hip; the disadvantages include the possible development of pain in the groin and acetabular erosion21,26. In a study of 166 patients who were managed for an acute fracture of the femoral neck with either a hemiarthroplasty or a total hip arthroplasty, Gebhard et al.10 found that the perioperative rate of complications was comparable for the two procedures but total hip arthroplasty resulted in better relief of pain and better function. Dorr et al.8, in a study of eighty-nine patients who had been randomized to treatment of an acute fracture of the femoral neck with either a total hip arthroplasty or a hemiarthroplasty with or without cement, found more perioperative complications (mostly dislocation) in the group that had a total hip arthroplasty. However, patients who had a high level of activity (those who walked about the community) had better relief of pain and better function after a total hip arthroplasty than did such patients after a hemiarthroplasty. Our findings are consistent with the results of those studies and of others5,11,12,31,34, and they show that a total hip arthroplasty effectively restores function and reliably relieves pain in most patients who have a fracture of the femoral neck. We found, as have other investigators2,4, that a subgroup of patients lost some function after the fracture and subsequent treatment. Advancing age; worsening medical problems; and, perhaps in some patients, limitations related directly to the arthroplasty probably all played a role in the increased limitation of activity of these few patients. Medical complications occurred frequently after the total hip arthroplasties in our series. These problems were probably related to the magnitude and the non-elective nature of the procedure, the age of the patients, and the medical comorbidities19,20,27,34. Despite the perioperative medical complications, the rate of perioperative morbidity was low, perhaps due in part to the selection factors in our non-randomized series: elderly patients who were very sick usually were managed with hemiarthroplasty rather than total hip arthroplasty. The most notable complication was the high rate of dislocation (10 per cent) in the first several years after the operation. A similar finding has been reported by several other authors7,34. The frequency of dislocation probably is related to the fact that elderly patients are less able to comply with precautions against dislocation, have an increased risk of falls, and have poorer soft tissues. Most (eleven) of the thirteen hips that had a dislocation had only one and were successfully treated with closed reduction. The present study shows that, despite a high rate of early complications, the durability of a total hip replacement used for the treatment of an acute fracture of the femoral neck in elderly patients was good. The durability probably was due in part to the advanced age of the patients (median age, seventy-five years) at the time of the operation and the lower level of activity. A low rate of revisions has been reported by most other authors after shorter durations of follow-up6,8,10,34. Delamarter and Moreland6 found no loose prostheses in twenty-two patients who were followed for a mean of 3.8 years after a total hip arthroplasty for the treatment of a fracture of the femoral neck. In a study by Gebhard et al.10, the rate of revision was 2.3 per cent a mean of fifty-six months after forty-four total hip arthroplasties performed with cement. Dorr et al.8 followed thirty-nine patients for two to four years after a total hip arthroplasty with cement and reported one revision for aseptic loosening. In a study of 122 patients followed for 1.5 to 5.5 years after a total hip arthroplasty with cement for the treatment of a fracture of the femoral neck, Taine and Armour34 reported that the rate of revision due to aseptic loosening was 3 per cent. The findings of those studies and of our study contrast with the findings of Greenough and Jones11, who reported that, of thirty-seven patients followed for a mean of fifty-six months after a total hip arthroplasty (most had had insertion of a Charnley-Müller prosthesis with cement), 49 per cent had a revision or were scheduled to have one. Unlike the other studies, the study by Greenough and Jones included only patients who were less than seventy years old. Thus, factors related to the selection of patients and the design of the prosthesis may have contributed to the high rate of failure. It is of note that four of the six patients in our series who had a revision were less than seventy years old at the time of the primary total hip arthroplasty. In conclusion, both total hip arthroplasty and hemiarthroplasty can produce satisfactory short-term results in older patients who have an acute displaced fracture of the femoral neck. Total hip arthroplasty is associated with a substantial rate of morbidity and dislocation, but it has the advantages of providing reliable relief of pain and restoration of function. When it is indicated, a total hip arthroplasty for the treatment of an acute fracture of the femoral neck in an elderly patient can provide excellent clinical results and is associated with good long-term survival of the prosthesis.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |