Secondary Logo

Journal Logo

Primary Total Knee Arthroplasty for Complex Distal Femur Fractures in Elderly Patients

Rosen, Andrew, L*; Strauss, Elton

Section Editor(s): Strauss, Elton MD

Clinical Orthopaedics and Related Research: August 2004 - Volume 425 - Issue - p 101-105
doi: 10.1097/01.blo.0000132466.65220.62
SECTION I: SYMPOSIUM: Geriatrics in Orthopaedics

Fractures of the distal femur in the geriatric population are associated with a high incidence of postoperative complications and poor results. Nonunion, loss of fixation, and malunion of these fractures occur with all types of treatment. The postoperative treatment of these patients demands a lengthy period of limited weightbearing that can increase the rate of medical complications. Our experience with these challenging fractures caused us to consider the use of a primary distal femur replacement total knee arthroplasty with the goals of elimination of fracture healing issues, early mobilization, and immediate weightbearing. Twenty-four distal femoral replacement knee arthroplasties were done from July 1998 to January 1999. Reviewed with a mean followup of 11 months, 17 patients (71%) resumed their preoperative level of ambulation. Knee range of motion averaged 1°– 103°. No major surgical or significant medical complications were experienced by these patients. Our experiences with this small number of patients have shown that an immediate arthroplasty offers many advantages over open reduction and internal fixation for geriatric patients with poor bone quality, preexisting degenerative joint disease, and medical problems.

From the *Insall Scott Kelly Institute for Orthopaedics and Sport Medicine, New York, NY; and the †Leni and Peter W. May Department of Orthopaedic Surgery, Mount Sinai Medical Center, New York, NY.

Correspondence to: Andrew L. Rosen, MD, Insall Scott Kelly Institute for Orthopaedics and Sports Medicine, 10 Union Square East, Suite 3M, New York, NY 10003. Phone: 212-844-6750; Fax: 212-844-6751; E-mail:

Guest Editor

Fractures of the distal femur in the geriatric population are challenging and complex because of poor bone quality, high levels of comminution, preexisting arthritis, and osteochondral damage at the time of injury. The goal of returning a patient to prefracture level of function often is difficult to achieve. Operative techniques that completely address problems with fixation of osteoporotic bone and articular cartilage damage still are incomplete.8 Open reduction and internal fixation of the femur can be a lengthy procedure with complex fixation and large levels of blood loss. We have reviewed the results of an alternative method of treating these complex fractures in elderly patients.

There have been many approaches to the distal femur fracture in the geriatric patient. Osteoporosis and osteopenia are formidable adversaries of surgical fracture fixation. Medical complications are common in this population, with studies reporting 7–22% mortality rates in the first year after distal femur fracture.4,5,11 Postoperative early mobilization and ambulation are important to prevent complications including deep venous thrombosis (DVT), pneumonia, and decubitus ulcers. Principles of early range of motion (ROM) and weightbearing are in sharp contradiction to the need for rigid immobilization to maintain fracture fixation for these patients. Operative treatment and conservative treatment of these elderly patients continue to show consistently poor results in a clinical study.2 The question that we addressed was whether fractures in this population should be treated differently than in younger patients.

Previous experiences with distal femur fractures in elderly patients have shown outcomes that correlated well with fracture types. Many patients with AO Type A7 diaphyseal fractures treated with intramedullary rod fixation maintained ambulation function postoperatively but most patients lost independent function. Most patients lost at least one level of ambulatory function. Decline from independent weightbearing to the use of canes and walkers was common. Results for the patients with Type C fractures were worse with most patients requiring permanent placement in skilled-care facilities. Almost all of these patients needed a walker, and many required additional surgery such as a knee replacement.

We changed our treatment protocol to include a new choice of joint replacement instead of reconstruction. A group of patients who were functional and who had Type C fractures were offered the choice of a primary arthroplasty or open reduction and internal fixation (ORIF). Our goal was to determine if primary arthroplasty could provide good functional results in elderly patients.

Back to Top | Article Outline


Between April 1998 and January 2003, all geriatric patients sustaining an intraarticular distal femur fracture were considered for treatment with a distal femoral replacement arthroplasty. Patients with a failed ORIF of the distal femur with nonunion and hardware failure also were included. Contraindications to this procedure included prefracture nonambulatory status and the presence of active infection. Twenty-four patients accepted, with written informed consent, a choice of treatment with prosthetic knee replacement.

Through evaluation of radiographs by the primary surgeon, one fracture was classified to be of the AO B2 type. All others were comminuted AO C-type fractures. The average patient age was 76 years with a range of 68–85 years. All patients had independent prefracture community ambulation. Five patients (63%) reported significant preexisting arthritic knee pain before fracture. All patients had a history of at least one significant medical problem, including diabetes mellitus, heart disease, hypertension, renal disease, or history of thromboembolism. Several patients had a history of multiple medical problems.

Preoperative treatment included the use of a splint, pain control, and medical evaluation and optimization. Full radiographic evaluation, including use of multiple radiographic views and CT imaging, was completed to assess the fracture pattern accurately.

All operations were done by the same attending surgeon. Anesthesia consisted of either spinal with epidural catheter (17 patients) or general anesthesia (seven patients). The patients were positioned in the supine position with a pad beneath the operative hip. A sterile tourniquet was placed on the upper thigh after preparation and draping. A longitudinal extensile midline anterior incision was used with extension more proximally than usually necessary for a primary TKR. Previous longitudinal incisions were reincised in patients with previous operative scars. A medial parapatellar retinacular incision was made. The fracture was examined and confirmed to be intraarticular and displaced in all patients.

Placing the components with the proper rotation and length to maintain joint kinematics and patellar orientation is the most difficult aspect of using a distal femoral replacement prosthesis for distal femur fractures. By reducing the fracture fragments to a close approximation of their original anatomic position, the length of the femur was measured with a length of suture. The anterior aspect of the femur was marked with a longitudinal line more proximally to give a guide for rotation of the prosthesis. In patients in whom reasonable assessment of rotation was not possible, the linea aspera was used to make an approximation of the posterior aspect of the femoral shaft.

The fracture fragments, menisci, and cruciate ligaments then were excised carefully to avoid neurovascular structures. The minimum length of intact bone proximal to the fracture to be resected was determined with rough comparison with available modular implant sizes. An effort was made to preserve proximal bone initially, with additional resection made, after use of trial implants. A cerclage cable was placed prophylactically on one femur with a small, nondisplaced crack that extended more proximally. A transverse femur cut was made, and the femoral canal was reamed and prepared for the intramedullary stem of the implant.

The Stryker-Howmedica-OsteonicsTM (Rutherford, NJ) Modular Reconstruction System (MRS) rotating hinge total knee prosthesis was used in all but one patient. The LinkTM Endo-Model Rotational Knee system (Hamburg, Germany) rotating hinge total knee system was used for one patient for whom the Howmedica system was unavailable.

The tibia cut then was made using standard extramedullary alignment guides, and the tibia was prepared for tray insertion. The patella was examined and prepared for resurfacing in patients who had extensive cartilage damage or arthritis. If minimal patellar chondral disease was present, it was left unresurfaced.

Insertion of trial components allowed examination of patellofemoral articulation, and careful adjustment was made to rotation and length before cementing of all implants. The femoral component was cemented using a cement restrictor and pressurized cement-gun technique. The first six patients had a cemented all PE tibial component implanted. For the remaining 17 patients, a cemented long-stem modular tibial tray with a modular PE liner was used. This was changed when the HowmedicaTM system added the option of stemmed metal implants.

Wound closure was obtained using interrupted sutures over Jackson-Pratt drains. The vastus medialis was advanced proximally to improve patellar tracking. Routine total knee protocols standardized at our institution were used postoperatively. These included prophylactic antibiotics, continuous passive motion, DVT prophylaxis, and physical and occupational therapy consults. Patients were mobilized on the first postoperative day and began full weightbearing immediately.

The patients were followed up postoperatively using standard TKR protocol. All followups were done by the primary surgeon. Physical examinations were done to evaluate wound healing, ROM and gait. Functional status was evaluated by patient history and notes from the physical therapy centers. Interval radiographic examinations were done on each patient and evaluated by the primary surgeon. All data were compiled retrospectively from review of the patients’ hospital and office charts.

Back to Top | Article Outline


Operative time for the reconstructive procedure averaged 3.3 hours with a range of 2.0–4.8 hours. No intraoperative complications occurred. Postoperatively, our patients required an average of 2.5 units of allogenic blood transfusions, with two patients not requiring any transfusion.

The length of followup averaged 11 months with a range from 5–23 months. All of our patients regained full weightbearing and ambulatory abilities. Seventeen patients (71%) regained their full prefracture level of ambulatory aid requirement. Seven patients required an additional level of assistive device for ambulation. Some of these patients required no assistive devices before fracture and required a cane after arthroplasty. Others who previously required a cane needed a walker. Range of motion postoperatively averaged 102° (range, 1°– 103°).

One superficial wound infection occurred in our study group. The patient was treated with intravenous and oral antibiotics and did not require operative intervention. No deep infections were recognized, and no secondary procedures were necessary for wound complications.

One postoperative mechanical complication occurred. A patient disengaged the hinge mechanism of their total joint prosthesis during a fall. This occurred in the patient in whom a LinkTM prosthesis was used. An open reduction of the hinge post was required, and no additional episodes occurred. No clinically or radiologically recognized early loosening occurred in our study group.

No major medical complications occurred postoperatively. One patient had a transient episode of mental status changes that resulted in an aggressive but inconclusive medical evaluation. The patient’s symptoms cleared with only supportive therapy. One patient died of unrelated cardiac issues 13 months after surgery.

Back to Top | Article Outline


Supracondylar fractures, especially intracondylar types, are difficult to treat. Osteoporotic bone associated with the geriatric patient makes this fracture difficult to treat. Muscular forces cause shortening, posterior angulation, and condylar separation. Goals of restoration of stability and prefracture function are not often attained with either open or closed treatment of these fractures.

Numerous techniques and devices have been developed to address these fractures, but all have similarly mixed results. A 25% rate of hardware failure and reoperation are reported as the expected results of ORIF of fractures in geriatric patients.2 Addition of bone cement to the internal fixation construct may result in less hardware failures but still is associated with an 18% reoperation rate for periimplant fractures.11 Extraarticular fractures treated with an intramedullary nail have an 10–14% rate of nonunion in clinical studies.3,6 Our study of primary arthroplasty showed no early hardware failure or periprosthetic fractures. Newer techniques using locking plate devices may achieve improved union percentages, but the patient still requires a lengthy period of limited weightbearing.9

Preexisting knee arthrosis is a major factor that limits the results of treatment of distal femur fractures in geriatric patients. Most patients in this population have some level of underlying cartilage degeneration. Even the most successful surgical restoration of the joint surface may result in a stiff, painful knee postoperatively. After fracture healing is complete, many of these patients will require a total joint arthroplasty made complicated and difficult with distorted anatomy, joint contracture, and internal fixation devices present. Delayed arthroplasty, although technically a primary knee replacement, may require significant surgical exposure, extensor mechanism reconstruction, and bulk allografts. The option of replacing the joint initially, with one surgery, is a very attractive concept for patients with significant preexisting arthrosis.

Major medical complications unfortunately are common occurrences after all distal femur fractures. One study showed a 25% rate of serious medical events with mixture of closed and open treatments.2 The mortality rate within 1 year of treatment of these fractures is significant: 22% in a study of mixed treatments,2 and 30% in a study of ORIF.3 Early mobilization and ambulation are universal postoperative approaches to preventing these complications. Unfortunately, most treatment of distal femur fractures requires long periods of either nonweightbearing or partial weightbearing to achieve fracture union. For the elderly patient with decreased strength and coordination, nonweightbearing necessitates months of confinement to a bed or wheelchair. Our use of primary arthroplasty allowed an aggressive postoperative rehabilitation program with immediate weightbearing that showed no postoperative major medical declines or mortalities.

The dilemma of function and fixation is very similar to issues involved in treatment of proximal femur fractures. Because of the high rate of nonunion seen with displaced femoral neck fractures, the technique of hip hemiarthroplasty is recognized as the gold standard for displaced fractures in geriatric patients. Patients with displaced distal femur fractures now have a similar option.

These types of rotating hinge TKRs have been used for many years by tumor surgeons for limb-salvage reconstruction after distal femoral resection. This implant design allows tibial rotation to decrease stress on the bone-cement interface while still maintaining stability for ambulation. Newer, second-generation hinge designs have shown good clinical results.1 The HowmedicaTM device seems to have lower rates of loosening and complications than other designs.10 Although some studies have shown high rates of revision needed for these components, most of these implants have been placed in younger patients with a resection of primary bone lesions. Elderly patients with low physical demands have a lower rate of wear-related failure of components.

The cost effectiveness of this technique is a concern. Although the surgical time is slightly longer than for a complicated open reduction, the implant costs are much higher. Future studies will need to examine the financial aspects of lower levels of nursing care required and the decreased risk of complications and additional procedures. It is likely that the additional cost of implants will be more than offset by these potentially cost-saving benefits of arthroplasty.

Our study was done as a retrospective analysis of a select group of patients that we thought would benefit from this new technique. As with any retrospective analysis, a lack of a control group limits the ability to directly compare the results of this technique with results of open reduction and internal fixation. However, our study does give validity to the use of arthroplasty in describing results that show fewer complications than many studies examining internal fixation.

Our choice of patients for inclusion in the study may have affected the validity of the study as well. All patients who met our inclusion criteria were offered a choice of treatment with primary arthroplasty or internal fixation. When informed of the potential benefit of arthroplasty, the majority of patients elected to proceed with primary knee replacement. The few patients who chose internal fixation and were not included could introduce some minimal selection bias into our results.

A final limitation of our study reflects the followups. All the patient results were observed and recorded by the primary surgeon. Early followup data were verified independently from the physical therapists’ evaluations, however, this was not possible once the patients discontinued therapy. Complete independent observation of each patient’s progress would have been more valid, but was not able to be done.

Despite some limitations, our early results show that distal femoral replacement arthroplasty can be a successful technique for limiting complications in the treatment of distal femur fractures in certain geriatric patients. Our relative indications include intraarticular fractures, severe articular damage, preexisting arthrosis, severe osteopenia, and malunions or nonunions.

For patients with severe osteopenia, preexisting arthrosis, restricted lifestyles, and limited treatment expectations, primary arthroplasty seems to be an appropriate technique. These procedures can achieve good functional results while minimizing medical morbidity and mortality.









Back to Top | Article Outline


1. Barrack RL, Lyons TR, Ingraham RQ, et al. The use of a modular rotating hinge component in salvage revision total knee arthroplasty. J Arthroplasty. 2000;15:858–866.
2. Butt MS, Krikler SJ, Ali MS. Displaced fractures of the distal femur in elderly patients: Operative versus non-operative treatment. J Bone Joint Surg. 1996;78B:110–114.
3. Dunlop DG, Brenkel IJ. The supracondylar intramedullary nail in elderly patients with distal femoral fractures. Injury. 1999;30:475–484.
4. Gynning JB, Hansen D. Treatment of distal femoral fractures with intramedullary supracondylar nails in elderly patients. Injury. 1999;30:43–46.
5. Karpman RR, Del Mar NB. Supracondylar femoral fractures in the frail elderly: Fractures in need of treatment. Clin Orthop. 1995;316:21–24.
6. Moskal JT, O’Shea JJ. Intramedullary stabilization of distal-third femur fractures in octogenarians. J Orthop Assoc. 2003;12:3–8.
7. Rüedi TP, Murphy WM. AO Principles of Fracture Management. Stuttgart, Thieme 2000.
8. Schatzker J. Fractures of the distal femur revisited. Clin Orthop. 1998;347:43–56.
9. Schutz M, Muller M, Krettek C, et al. Minimally invasive fracture stabilization of distal femoral fractures with the LISS: a prospective multicenter study: Results of a clinical study with special emphasis on difficult cases. Injury. 2001;32(Suppl):SC48–SC54.
10. Shih LY, Sim FH, Pritchard DJ, et al. Segmental total knee arthroplasty after distal femoral resection for tumor. Clin Orthop. 1993;292:269–281.
11. Struhl S, Szporn MN, Cobelli NJ, et al. Cemented internal fixation for supracondylar femur fractures in osteoporotic patients. J Orthop Trauma. 1990;4:151–157.
© 2004 Lippincott Williams & Wilkins, Inc.