 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1998 The Journal of Bone and Joint Surgery, Inc. Postoperative Weight-Bearing after a Fracture of the Femoral Neck or an Intertrochanteric Fracture* , NEW YORK, N.Y.Investigation performed at the Department of Orthopaedic Surgery, Hospital for Joint Diseases, New York City
Sixty patients who had had operative treatment of a fracture of the femoral neck or an intertrochanteric fracture were allowed to bear weight as tolerated on the injured limb. The average age was seventy-seven years. Computerized gait-testing was performed at one, two, three, six, and twelve weeks postoperatively to quantify weight-bearing. For the purpose of analysis, the patients were divided into three groups according to whether they had internal fixation of a stable fracture, internal fixation of an unstable fracture, or a primary hemiarthroplasty. Thirty-two patients completed the entire twelve-week study. The average amount of weight that these patients placed on the injured limb increased progressively with time. The average load supported by the injured limb was 51 per cent that of the uninjured limb at one week, and it gradually increased to 87 per cent at twelve weeks. During the first three weeks, the patients who had had internal fixation bore substantially less weight than those who had had a hemiarthroplasty. By six weeks, we could detect no significant differences, with the numbers available, among the groups with regard to weight-bearing or other measured gait parameters. We concluded that elderly patients who are allowed to bear weight as tolerated after operative treatment of a fracture of the femoral neck or an intertrochanteric fracture appear to voluntarily limit loading of the injured limb.
The preferred weight-bearing regimen after operative treatment of a fracture of the femoral neck or an intertrochanteric fracture remains controversial. Restricted weight-bearing can delay the functional recovery of elderly patients as well as their return to independent living8,13,14; however, some surgeons are hesitant to permit early, unrestricted weight-bearing. It has been demonstrated that unrestricted weight-bearing does not increase the rate of complications after internal fixation of non-displaced fractures of the femoral neck and stable intertrochanteric fractures1-3. Patients who have a displaced fracture, however, often are restricted to partial weight-bearing because of the potential for failure of fixation and the risk of non-union4,9,10. It has been the practice at our hospital to allow all patients to bear weight on the injured limb as tolerated after operative treatment of a fracture of the femoral neckor an intertrochanteric fracture. We believe that patients will voluntarily limit weight-bearing on the basis of the degree of discomfort or apprehension that such weight-bearing causes. We performed a study of patients who were allowed to bear weight as tolerated after operative treatment of a fracture of the femoral neck or an intertrochanteric fracture. The purposes of the investigation were to quantify weight-bearing and other gait parameters in the early postoperative period and to determine if weight-bearing was related to the type of fracture and treatment.
Between January 1, 1995, and September 15, 1996, seventy-one consecutive patients who were more than fifty years old were managed operatively by the senior one of us (K. J. K.) for a fracture of the femoral neck or an intertrochanteric fracture at the Hospital for Joint Diseases. All of the patients were considered for inclusion in the present study, which was approved by the Institutional Review Board. Exclusion criteria included dementia, disease involving the contralateral hip, or a fracture of the ipsilateral lower extremity that precluded unrestricted weight-bearing. Sixty patients (forty-four women and sixteen men) met the inclusion criteria and consented to participate in the study. The average age was seventy-seven years (range, fifty-four to ninety-two years). The type of operative treatment was decided by the operating surgeon on the basis of the time since the injury, the amount of comminution of the fracture, and the general health of the patient. The patients were divided into three groups on the basis of the type of fracture and treatment. The first group consisted of seven patients who had a stable, non-displaced fracture of the femoral neck (Garden type I or II) that was stabilized with multiple, cannulated, cancellous-bone lag screws and fifteen patients who had a stable intertrochanteric fracture that was stabilized with a sliding hip screw. The second group consisted of eleven patients who had an unstable, displaced fracture of the femoral neck (Garden type III or IV) that was reduced and stabilized with multiple, cannulated, cancellous-bone lag screws and twelve patients who had an unstable intertrochanteric fracture that was stabilized with a sliding hip screw. The third group consisted of fifteen patients who had a displaced fracture of the femoral neck that was treated with a unipolar hemiprosthetic hip replacement inserted with cement. Closed reduction and internal fixation was attempted for all displaced fractures of the femoral neck, and hemiarthroplasty was performed when this could not be achieved. Intertrochanteric fractures were defined as stable when an intact posteromedial cortical buttress was seen radiographically (two-part fractures), and they were considered unstable when there was loss of a posteromedial cortical buttress (fractures with three parts or more). All patients followed a similar postoperative protocol, which consisted of standing with assistance, with full unrestricted weight-bearing and use of an assistive device, as tolerated, on the first postoperative day. As the walking ability improved, the assistive device was changed as determined appropriate by a physical therapist. Computerized weight-bearing and gait analysis was performed at one, two, three, six, and twelve weeks postoperatively with use of the F-Scan System (Tekscan, Boston, Massachusetts). With this system, a 900-element-array flexible flat transducer (0.2 millimeter thick) is inserted between the foot and the footwear to measure plantar loads. The sensors were cut to the size of the feet, and all transducers were calibrated before each test. Standardized footwear (lightweight thin-soled shoes) was worn by all patients. Before the weight-bearing was recorded, each patient walked twenty-five feet (7.6 meters) with the transducers in place. Data-recording was then initiated after the patient had walked several additional steps to ensure a representative recording. From a forty-five-second gait-recording of approximately seven to ten steps, five representative steps by each foot were used to determine the average load (defined as the integrated plantar load from heel-strike to toe-off divided by the time interval), maximum load, and time per step for both the injured and the uninjured limb with use of the F-Scan software. The total time was measured as the time from heel-strike of one foot to the next heel-strike of the same foot. For each of the load variables, the value for the injured limb was calculated as a percentage of that for the uninjured limb. Thirty-two of the sixty patients completed the entire twelve-week testing protocol. Some patients were dropped from the study at six and twelve weeks because scheduling difficulties had caused them to miss the gait tests; many of these patients lived in a nursing home, and no financial compensation was provided for participation or transportation. Of the thirty-two patients who remained in the study, twenty-five were women and seven were men. The average age was seventy-five years (range, fifty-four to eighty-seven years). Of the thirty-two fractures, eleven were stable (five involved the femoral neck and six were intertrochanteric), fifteen were unstable and treated with internal fixation (nine involved the femoral neck and six were intertrochanteric), and six were displaced and treated with a hemiarthroplasty.
Statistical Analysis
One week postoperatively, twenty-four patients (75 per cent) used a walker, five (16 per cent) used crutches, and three (9 per cent) used a cane. By twelve weeks postoperatively, four patients (13 per cent) used a walker, one (3 per cent) used crutches, seventeen (53 per cent) used a cane, and ten (31 per cent) walked without any assistive device. The average load supported by the injured limb, recorded as a percentage of that supported by the uninjured limb, increased significantly with time for all thirty-two patients—that is, the loads at one, two, and three weeks were less than those at six and twelve weeks (typical p = 0.0001, effect size = 0.473, power = 1.00). The average load was 51 per cent at one week, 64 per cent at two weeks, 65 per cent at three weeks, 80 per cent at six weeks, and 87 per cent at twelve weeks. Either the patients who had had a stable fracture or those who had had a hemiarthroplasty had the highest average load at each of the five time-periods (Fig. 1). At one, two, and three weeks, the patients who had had a hemiarthroplasty bore a greater average load than the patients in the other two groups. By six weeks, however, we could detect no significant differences, with the numbers available, among the three groups with regard to average load (p = 0.16, effect size = 0.120, power = 0.375).
The average maximum load supported by the injured limb, recorded as a percentage of that supported by the uninjured limb, also increased with time. It was 69 per cent at one week, 78 per cent at two weeks, 80 per cent at three weeks, 88 per cent at six weeks, and 91 per cent at twelve weeks. The patients who had had a hemiarthroplasty bore a significantly greater maximum load in the first three weeks postoperatively than did those who had had internal fixation (p = 0.035, effect size = 0.130, power = 0.836). By six weeks, however, we could detect no significant differences among the three groups, with the numbers available (p = 0.245, effect size = 0.081, power = 0.37). In the entire series of thirty-two patients, the average duration of the gait cycle for the injured limb, recorded as a percentage of that for the uninjured limb, was approximately 90 per cent at one, two, and three weeks; 93 per cent at six weeks; and 97 per cent at twelve weeks. At each time-period, we could detect no significant differences among the three groups, with the numbers available (p = 0.19, effect size = 0.109, power = 0.341). The total duration of one gait cycle significantly decreased with time—that is, the cycle took longer at one, two, and three weeks than at six and twelve weeks, and it took more time at one week than at two and three weeks (typical p = 0.0001, effect size = 0.471, power = 1.00). The average time for one gait cycle was 5.8 seconds at one week, 4.5 seconds at two weeks, 3.8 seconds at three weeks, 2.6 seconds at six weeks, and 2.4 seconds at twelve weeks (Fig. 2). We could detect no significant differences among the three groups at any time-period, with the numbers available (p = 0.364, effect size = 0.067, power = 0.215).
Fifty-eight of the original sixty patients had uneventful healing at twelve weeks. Two patients (who did not complete the entire testing protocol) who had a displaced fracture of the femoral neck had complications related to healing. One of them, a seventy-nine-year-old woman, had a loss of fixation resulting in one centimeter of displacement three weeks postoperatively, but she did not want an additional operation. The average weight-bearing loads were approximately 30 per cent less after the failure of the fixation. The other patient, a seventy-three-year-old man, had a non-union of the femoral neck, and a total hip replacement was performed eight months after the fracture. The initial weight-bearing loads were high for two weeks and then decreased with subsequent measurements. A comparison of the radiographs made at the most recent follow-up examination with those made postoperatively revealed no gross translation of the fracture (displacement of more than five millimeters) or deformity (varus angulation of more than 5 degrees) in the thirty-two patients who completed the study.
The results of the present study are expressed as percentages of the loads borne by the uninjured limb. Because the maximum load on the uninjured limb is approximately 1.2 times body weight11, the actual load on the injured limb in the initial three weeks postoperatively can be calculated as 60 to 80 per cent of body weight. These loads are important factors in the design of fixation devices and essential factors in biomechanical analyses. Assuming that the elderly patients did not bear any appreciable weight on the walker, crutches, or cane, loads of two to three and one-half times body weight on the hip and fixation device can be expected during gait5,7,11. In comparison, non-walking activities, such as moving around in bed and using a bed pan, can generate loads of three to five times body weight5,11 on the injured hip as a result of muscle contraction. As only thirty-two patients completed the entire study, the groups for the comparative analysis were small. A repeated-measures analysis-of-variance design rather than a randomized design was used because fewer patients are needed for statistical power. Analysis of the same patients at the same time-periods removes the error due to the contribution of individual differences among patients. However, because the groups are small, the results of the statistical comparisons can be misleading. Other limitations of this study include the short-term follow-up and the fact that data were available at twelve weeks for only 53 per cent of the initial study group. In addition, it was difficult to obtain data for early weight-bearing because some of the patients were not able to walk well enough for testing until two weeks postoperatively. It is possible that some of these elderly patients became fatigued during testing, resulting in altered gait and weight-bearing7, but this was not observed. In addition, use of an assistive device (such as a cane, crutches, or a walker) was not standardized. The type of device that the patient used was determined by a physical therapist and was changed as the ability to walk improved. It is our usual clinical practice, however, to allow patients to progress from use of a walker as tolerated. Although each type of assistive device may have a different effect on weight-bearing load, no obvious relationship between the type of assistive device and the weight-bearing load was noted in any of the three groups. Finally, although we noted no major radiographic displacement or deformity of the fracture, minor changes could have affected gait and resultant weight-bearing. To the best of our knowledge, this is the first study to quantify weight-bearing by patients who have had operative treatment of a fracture of the femoral neck or an intertrochanteric fracture. These patients, who were allowed to bear weight as tolerated, bore 60 to 80 per cent of body weight on the injured limb in the early postoperative period. The patients in whom a stable or unstable fracture had been treated with internal fixation bore substantially less weight than those who had had a hemiarthroplasty. We concluded that patients who are allowed to bear weight as tolerated after treatment of a fracture of the femoral neck or an intertrochanteric fracture appear to voluntarily limit loading of the injured limb.
*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. |