Management of Fractures Around Total Knee Replacement

by Jerome D. Wiedel, M.D.

In: Total Knee Arthroplasty: A Comprehensive Approach
Editors: Hungerford, David S., Krackow, Kenneth A. and Kenna, Robert V.

Fractures about replacement components are becoming more widely recognized as a complication following total knee arthroplasty. The incidence seems to be increasing as the number of patients and the length of follow-up increase. Several authors have reported on this complication (1-4, 6-8, 10, 11) with two series recently describing some of the causes and methods of treatment (5, 9).

Fractures around a total knee replacement may occur at three different sites: distal femur, proximal tibia, and patella. In this chapter fractures around total knee replacements are discussed under each of these anatomic sites emphasizing etiologic factors, mechanisms of injury, methods of treatment, and complications.

Distal Femur Fractures

The mechanism of injury causing the distal femur fracture invariably involves minor trauma, usually tripping or slipping and falling to the floor. Such minimal force as twisting the leg, in either a standing or recumbent position, has caused this fracture.

In reviewing reported series and our own cases, several factors may account for the ability of such trivial trauma to cause these fractures. Osteoporosis is probably the most consistent finding. Furthermore, many of these patients preoperatively were either nonambulatory or minimally so, and once an initially successful total knee replacement has been performed they have become active. However, because of unsteadiness and residual weakness, they are more susceptible to injury.

In reviewing the types of prostheses associated with fractures, it was apparent that prostheses which required more bone resection, particularly removal of the posterior femoral cortex, had a greater risk for this complication.

The type of distal femoral fractures following total knee replacement are quite similar to the recognized standard classification: 1) distal shaft, 2) supracondylar, and 3) condylar. These will be discussed separately since the management and complications may differ significantly. It should be noted that this material represents a review of the author's own series of cases, a number of which have at sometime been treated by different attending orthopaedists.

Distal Femoral Shaft Fractures

A distal femoral shaft fracture may occur clearly proximal to the femoral component (Fig. 15.1), seemingly unrelated to the prothesis or at the proximal extent of a long stemmed prosthesis (Fig. 15.2). Fractures occurring about a stemmed prosthesis are likely to be oblique, indicating a torsional stress-induced fracture. A fixed-hinged prosthesis may further contribute to the likelihood of this type of fracture. With no axial rotation possible within the prosthesis, increased torsional stress is undoubtedly generated within adjacent bone.

Management of these fractures was by closed means; initially, skeletal traction and balanced suspension followed by external support in the form of a brace or splint. All the fractures united in satisfactory position. Three patients' postinjury status remained unchanged from their preinjury status. One patient (St. Georg prosthesis) remained in a wheelchair after fracture healing with a limited range of motion (65-90°).

Figure 15.1. Distal femoral shaft above a nonstemmed femoral component.

Supracandylar Fractures

This group of fractures represents the most common type following total knee replacement. It is the most commonly reported type and, in the author's series, accounted for 11 of the 17 fractures. Most of these fractures are oblique with varying degrees of comminution and occasionally with condylar involvement. The fracture deformity, when present, was interestingly the opposite of that typically described for a supracondylar fracture (Fig. 15.3). This finding may be the result of a different mechanism of injury in that these patients usually fall onto their knees, forcing the distal fragment with the prosthesis posteriorly. Furthermore, the relative muscular weakness in these patients results in minimal, if any, deforming force with which to produce the typical deformity.

All of the fractures were closed except for one which had a small anterior laceration at the fraction level. This fracture initially had an open debridement followed by traction. Eight other fractures were also initially managed by closed means, either in traction or plaster immobilization. The two remaining fractures in this group were initially managed open. One had an open reduction and blade plate fixation. The prosthesis present was a Marmor bicompartmental replacement. The other case was a short oblique fracture in a patient with a Spherocentric prosthesis which was revised to a long stem Guepar prosthesis (Fig. 15.4, A and B).

The prostheses present in this group of fractures included: Spherocentric, 5; UCI, 3; total condylar, 1; Marmor, 1; and Walldius, 1.

The results of treatment included three delayed unions and one nonunion. One delayed union occurred in the patient previously described who had an open fracture above a total condylar prosthesis. An open procedure with blade plate fixation was accomplished 8 weeks after the fracture, and the fracture ultimately healed.

The other two cases of delayed union occurred in patients with Spherocentric prostheses. One eventually healed after 6 months in plaster, and the other one was revised to a long stem prosthesis after 4 months of closed management failed to gain union.

The case of nonunion occurred in a short oblique fracture above a Spherocentric prosthesis. After 4 months of closed management, union was not evident (Fig. 15.5A); therefore, revision to a long stem Spherocentric prosthesis was performed. Fifteen months later nonunion persisted (Fig. 15.5B), at which time bone graft and an osteostimulator were inserted (Fig. 15.5C). Thirty-two months after fracture the patient died of causes unrelated to the total knee replacement. The final status of union was never determined. The patient was able to transfer and ambulate minimally, using a walker.

Figure 15.2. Oblique distal femoral shaft fracture at the level of a stemmed femoral component.

Malunion occurred in two cases, one with a Marmor prosthesis and one with a Spherocentric. Of the remaining seven fractures, one patient died before completion of healing, and six healed in satisfactory position (Fig. 15.6, A and B).

The Spherocentric unit was the single most common prosthesis involved with fractures in this series (5), and all were of the supracondylar type. In these 5 cases different circumstances led to different solutions. One patient had a loosened tibial component prior to the fracture. One patient underwent primary revision to a long stem prosthesis, and one had a revision to a long stem prosthesis after the fracture was considered a delayed union 4 months following closed management. A fourth case went on to a malunion and loosening of the femoral component (Fig. 15.7) and the fifth developed a nonunion (previously described).

In review of the supracondylar fractures it is quite apparent that significant complications can occur. There is not one form of management that will give uniformly good results. Each case has to be individually considered on the basis of the patient's status (medical, ambulatory), the characteristics of the fracture (degree of comminution, deformity), and the type of prosthesis present.

A fracture amenable to internal fixation which would not interfere with the prosthesis may be the ideal management to avoid prolonged bed rest. However, many of these fractures are to comminuted for internal fixation, and the prosthesis may be of the type that would not allow use of conventional internal fixation devices. An alternative in this case would be to revise the prosthesis, using a longer stemmed device to stabilize the fracture.

The supracondylar fractures occurring around long stemmed prostheses have internal fixation already in place and may be amenable to closed management with early ambulation, using minimal external support.

The standard short stemmed Spherocentric prosthesis presents a particularly difficult management problem. The type of treatment- whether open, in which case it would probably have to be a revision of the prosthesis, or closed-must be considered on an individual basis. In situations where the surgeon has strong preference for a Spherocentric unit, review of our experience suggests that use of a longer stemmed unit, rather than the standard short stemmed device, may lessen the chance of subsequent supracondylar fracture.

Figure 15.3. Drawing demonstrates typical deformity of a supracondylar femur fracture and a radiograph demonstrates the opposite deformity in a supracondylar femur fracture above a Spherocentric prosthesis.

Figure 15.4. (A) Displaced oblique supracondylar femur fracture above a Spherocentric prosthesis. (B) Twelve months after revision to a long stemmed femoral component.

Condylar Fractures

Condylar fractures occur only in the presence of nonstemmed prostheses. In the author's series, two cases were encountered; one in a geometric and the other a total condylar prosthesis (Figs. 15.8 and 15.9). Both presented with posteriorly displaced and angulated distal fragments and prostheses. In spite of traction and attempts at closed manipulative reduction under anesthesia, satisfactory reduction could not be obtained. Open reduction and internal fixation were not felt feasible. Therefore, the deformity was allowed to persist, managing both cases closed. These fractures healed but in unsatisfactory positions (Fig. 15.10, A and B). The geometric prosthesis shown in Figure 15.10A was later revised to a Spherocentric unit. At surgery the femoral component was noted to be stable; however, the tibial component was loose.

The other case has an unsatisfactory range of motion (0-50°), and this patient requires two crutches and a brace for walking.

Figure 15.5. (A) Nonunion of supracondylar femur fracture above a Spherocentric prosthesis. (B) Fifteen months after revision nonunion persisted. (C) After bone graft and osteostimulator implanted.

Figure 15.6. (A) Comminuted supracondylar femur fracture above a UCI prosthesis. (B) Five years later, fracture united, range of motion 0-100°.

Figure 15.7. Malunion of supracondylar femur fracture with loosened Spherocentric femoral component.

Figure 15.8. Condylar fracture above a geometric prosthesis.

Figure 15.9. Condylar fracture above a total condylar prosthesis.

Figure 15.10. (A) Malunion of condylar fracture shown in Figure 15.8. (B) Malunion of condylar fracture shown in Figure 15.9.

Conclusions-Femoral Fractures

The following conclusions were suggested after review of distal femur fractures around total knee replacement:

  1. Distal femoral shaft fractures are the most amenable to closed management and have the lowest complication rate.
  2. Suprarondylar and condylar fractures have a significant complication rate.
  3. The type of prosthesis will influence the management of supracondylar fractures.
  4. Fracture deformity in supracondylar and condylar fractures is frequently the opposite of the typical deformity observed in similar fractures without prostheses.
  5. Fracture healing does not seem to be impaired by the presence of prothesis or cement.
  6. Closed treatment is recommended unless the reduction is unsatisfactory. In this case either open reduction and internal fixation should be done, providing the fracture is amenable to fixation, or primary prosthesis revision to a long stemmed femoral component may be performed.

Proximal Tibial Fractures

Proximal tibial fractures following total knee replacement occurring as a result of trauma are rare. In the author's review of over 800 cases which produced 17 cases of distal femur fractures, no case of traumatically induced tibial fracture was encountered. The type of proximal tibial fracture encountered following total knee replacement is typically a stress fracture (Fig. 15.11). These fractures are observed in cases where axial malalignment has produced abnormally high, eccentric loads on either the lateral or medial tibial plateau. Medial tibial plateau fractures with associated varus malalignment are the most commonly observed stress fractures (2, 9).

A loosened tibial component has been observed in all reported cases and has been present in all cases reviewed by the author. The tibial plateau fractures associated with a loosened tibial component and axial malalignment are best managed by revision arthoplasty. It is critical to reestablish the normal physiologic alignment of 5-10° valgus to prevent this complication.

In nonoperative candidates, external bracing using a hinged brace may provide some relief of patient's symptoms and aid in their ambulation.

Figure 15.11. Fracture of medial tibal plateau. Tibial component is loose and genu sarus deformitynoted.

Patella Fractures

Patellar fractures following total knee replacement are rare indeed and have only become a concern since prosthetic patellar replacement has become a routine part of total knee procedures. The cases reviewed by the author have occurred as a result of a fall or simply upon getting up from a sitting position.

Etiologic factors are osteoporosis and a weakened patella as a result of resection for patellar replacement. Technical errors in removing too touch bone or malpositioning the prothesis may also contribute to the causes of patellar fracture.

The management of these fractures is essentially the same as for patellar fractures without associated prosthetic arthoplasty. If the fracture is minimally displaced or nondisplaced, conservative treatment is recommended. Significantly displaced fractures with disruption of the extensor mechanism should be operated upon if possible. The actual procedure performed will depend upon the condition of the bone.

If the fracture is amenable to fixation and the prosthesis is not loose, simply fixing the fracture should be considered. However, if the prosthesis is loose, a decision must be made as to whether the fracture can be fixed, followed by reinsertion of a prosthesis, or whether patellectomy may not be the best solution (Fig. 15.12).

Figure 15.12. Displaced patella fracture treated with patellectomy.


Fractures about the knee following total knee replacement will undoubtedly increase as the number of patients with total knee replacement and the length of follow-up increases.

Our main emphasis should be on prevention. This can be accomplished, at least in part, by proper selection of patients, proper selection of the prosthesis, and performing technically accurate procedures.

In the face of fracture, each case must be carefully evaluated and the best method of treatment selected on an individual basis. Factors to be considered in addition to the patient's medical and ambulatory status are the type of fracture and the type of prosthesis already in place.


  1. Bargeen JH, Freeman MAR, Swanson SAV, et at.: ICLH (Freeman/Swanson) arthroplasty in treatment of arthritic knee - a two- to four-year review. Clin Orthop 12:65-75, 1976.
  2. Bryan RS, Lowell FA, Peterson MD, et al.: Polycentric knee arthroplasty, a preliminary report of postoperative complications in 450 knees. Clin Orthop, 94:148-152, 1973.
  3. Engelbrecht E, Siegel A, Rottger J, et al.: Statistics of total knee replacement: partial and total knee replacement, design St Georg, a review of a four-year observation. Clin Orthop 120:54-63, 19976.
  4. Freeman PA: Walldius arthroplasty, a review of 80 cases. Clin Orthop 94:85-91, 1973.
  5. Hirsh DM, Bhalla S, Roffman M: Supracondylar fracture of the femur following total knee replacement, report of four cases. J Bone Joint Surg 63A:162-163, 1981.
  6. Jackson JP, Elson RA: Evaluation of the Walldius and other prostheses for knee arthroplasty. Clin Orthop 94:104-114, 1973.
  7. Jones GB: Total knee replacement-the Walldius hinge. Clin Orthop 94:50-57, 1973.
  8. Merryweather R, Jones GB: total knee replacement, the Walldius arthroplasty. Orthop Clin North Am 4:585-596, 1973.
  9. Rand JA, Coventry M: Stress fractures after total knee arthroplasty. J Bone Joint Surg 62A:226-233, 1980.
  10. Vanhegan JAD, Dabrowski W, Arden GP: A review of 100 Attenborough stabilized gliding knee prostheses. J Bone Joint Surg 61B:445-450, 1979.
  11. Watson JR, Wood H, Hill RCJ: The Shiers arthroplasty of the knee. J Bone Joint Surg 56B:300-304, 1974.