Abstract
Introduction
Periprosthetic distal femur fractures are severe injuries occurring in the often osteoporotic bone of the elderly. Far cortical locking (FCL) screws, which have been shown to promote increased callus formation in animal models, have recently become available for clinical use. The purpose of this study is to report preliminary healing and complication rates of periprosthetic distal femur fractures treated with FCL constructs.
Materials and Methods
A retrospective review of 20 patients who underwent open reduction and internal fixation of periprosthetic distal femur fractures using FCL constructs was performed. Healing was assessed radiographically and clinically at 6, 12 and 24 weeks post-operatively. Construct failure was defined as any hardware breakage or bone-implant dissociation leading to loss of reduction.
Results
Complete data through the 24 week study period was available for 18/20 patients. Bridging callus was identified in 16/18 patients by the 24 week follow up for a healing rate of 88.9%. In patients that healed, the average time to medial bridging callus formation was 10.7±6.7 weeks, 11.0±6.6 weeks for anterior fracture line and 13.4±7.5 weeks for the posterior fracture line. both patients that failed to heal underwent revision surgery.
Discussion
The initial results of this study are comparable to results reported for distal femur periprosthetic fractures treated with locking plate fixation without FCL screws, although it was difficult to compare time to healing between previously published studies. It is the impression of the authors that callus appears earlier and is more robust and uniform between the three cortices in FCL cases compared to their previous experiences with traditional locking plate periprosthetic distal femur fractures. This work suggests that FCL screws may be superior to traditional locking constructs but further studies are needed to directly compare the two methods.
Introduction
Periprosthetic distal femur fractures are severe injuries occurring in approximately 0.2% to 2.5% of all total knee replacements (TKR)1,2. These numbers can be expected to increase substantially due to the rising number of total knee arthroplasties performed in our aging population. Although many of the techniques are the same as those used for other distal femur fractures, the periprosthetic distal femur fracture presents unique challenges. The orthopaedic surgeon must evaluate the pre-injury functional status, bone quality, medical co-morbidities, prosthetic bone interface and assess possible points of fixation which are limited by the presence of the femoral prosthesis. The periprosthetic distal femur fracture is often comminuted and occurs in weak osteoporotic bone, making fixation particularly challenging. Bridging callus must form early and become mechanically stable to prevent fixation failure in these difficult fractures.
Locking plates are commonly used to definitively fix periprosthetic distal femur fractures. Compared to traditional plates, locking plates provide increased fixation stability in the often osteoporotic distal femur and also allow for minimally invasive insertion techniques3. However, clinical results of distal femur fractures treated with locking plates have indicated that healing has not been improved over other techniques3. The high stiffness of locking plates may decrease micro-motion at the fracture site, limiting callus formation and secondary fracture healing4.
Far cortical locking (FCL) screws have been shown to promote callus by providing a biomechanical environment and healing response for locking plates similar to that provided by external fixators4. Symmetric motion across the fracture site combined with decreased construct stiffness has been shown to increase callus formation compared to traditional locking plates. In an ovine fracture model, Bottlang et al5. reported FCL screw constructs increased callus volume by 36% over traditional locking plates. The callus also contained 44% higher bone mineral content and was 54% stronger in torsion5. Bone mineral content was also comparable between the near and far cortices in the FCL group, while the near cortex of the traditional locking plate group exhibited 49% less bone mineral content at the near cortex compared to the far cortex.
FCL screws are now available for clinical use but healing results and early complication rates have not been reported. It is important to carefully audit the results of new technology as it is released. The purpose of this study is to report preliminary healing and complication rates of periprosthetic distal femur fractures treated with far cortical locking constructs and compare these results to previously reported data for periprosthetic distal femur fractures fixed with traditional locking plates.
Materials and Methods
This is a retrospective case series that assesses 20 consecutive patients with periprosthetic distal femur fractures above a TKR that were treated from April 2011 – July 2012 with far cortical locking screws and plates at three orthopaedic centers: The University of Iowa Hospitals and Clinics in Iowa City, IA, Slocum Orthopedics and Sports Medicine Clinic in Eugene, Oregon and Legacy Emanuel Hospital and Health Center in Portland, Oregon. During the study period the surgeons exclusively used this type of locking screw for these fractures. Occasional fractures were treated non-operatively. Inclusion criteria included non-pathological distal periprosthetic femur fractures around stable TKR which were treated with open reduction and internal fixation using locked plating with FCL screws. Operation notes and medical records were reviewed for patient characteristics (age, gender, BMI, diabetes mellitus, and smoking), fracture characteristics (side, mechanism of injury, AO/OTA classification and open vs. closed fracture), implant detail, healing and complications (implant failure and non-healing). Healing was assessed radiographically by two surgeons (SM and ZR) on anteroposterior (AP) and lateral radiographs obtained at 6, 12 and 24 week post-operative clinic visits. As initially described by Whelan et al6, fractures were considered healed when bridging callus was identified separately for medial, anterior and posterior fracture lines in combination with clinical exam which demonstrated absence of pain and ability to tolerate progressive weightbearing. Construct failure was defined as any hardware breakage or bone-implant dissociation leading to loss of reduction. Serial radiographs were also reviewed for evidence of osteolysis and implant loosening.
Results
Patient demographics and fracture characteristics are summarized in Table 1. There were 19 females and 1 male with an average age of 77.0 years and average BMI of 31.4. Eighteen of the twenty fractures occurred as a result of a ground level fall, with the remaining two fractures resulting from a motor vehicle accident and a horse accident. Three of the fractures were open. Data on healing and complications are shown in Table 2. Complete data through the 24 week post-operative study window were available for eighteen of the twenty patients. Bridging callus was identified separately for the medial, anterior and posterior fracture lines in sixteen of the eighteen patients by the 24 week follow up for a healing rate of 88.9%. In patients that healed, the average time to medial bridging callus formation was 10.7±6.7 weeks, 11.0±6.6 weeks for anterior fracture line and 13.4±7.5 weeks for the posterior fracture line. Figure 1 shows the pre-operative and 24 week post-operative radiographs of a patient who formed bridging medial, anterior and posterior callus and returned to baseline activity level.
Table 1.
Patient Demographics and Fracture characteristics
| Mean age in years (range) | 77 (43 - 95) |
| Gender (%) | |
| Female | 19 (95) |
| Male | 1 (5) |
| Mean BMI in kg/m2 (range) | 31.4 (21.6 - 50.0) |
| Comorbidities (%) | |
| Diabetes mellitus | 5 (25) |
| Smoker | 3 (15.0) |
| Side affected (%) | |
| Right | 10 (50.0) |
| Left | 10 (50.0) |
| Mechanism of Injury (%) | |
| Fall | 18 (90.0) |
| Motor Vehicle Collision | 1 (5.0) |
| Horse Accident | 1 (5.0) |
| AO/OTA Classification (%) | |
| 33A1 | 6 (30.0) |
| 33A2 | 3 (15.0) |
| 33A3 | 11 (55.0) |
| Classification (open/closed) (%) | |
| Open | 3 (15.0) |
| Closed | 17 (85.0) |
Table 2.
Healing and complications
| Patient | Medial Callus (wks) | Anterior Callus (wks) | Posterior Callus (wks) | Complications |
|---|---|---|---|---|
| 1 | 6 | 12 | 24 | |
| 2 | 6 | 12 | 24 | |
| 3 | 12 | 6 | 12 | |
| 4 | 6 | 6 | 6 | |
| 5 | 6 | 6 | Withdrew | Lost to follow-up |
| 6 | 12 | 6 | 12 | |
| 7 | 24 | 24 | 24 | |
| 8 | 6 | 12 | 6 | |
| 9 | 12 | 12 | 12 | |
| 10 | 6 | 6 | 6 | |
| 11 | 6 | 6 | 12 | |
| 12 | - | - | - | Continued pain with lack of bridging callus treated with revision fixation with bone grafting at 12 wks |
| 13 | 6 | 6 | 6 | |
| 14 | 12 | 12 | 12 | |
| 15 | 24 | 24 | 24 | |
| 16 | - | - | - | Lost to follow-up |
| 17 | 24 | - | - | Hardware failure at 9 months requiring revision surgery |
| 18 | 6 | 6 | 6 | |
| 19 | 6 | 6 | 6 | |
| 20 | 12 | 12 | 12 | |
| Avg±STDev | 10.7±6.7 | 11.0±6.6 | 13.4±7.5 |
Figure 1. Pre-operative and 24 week post-operative radiographs of an 82 year old diabetic female who sustained a closed 33A1 periprosthetic distal femur fracture following a ground-level fall. she was treated with locking plate fixation with proximal FcL screws. bridging callus is seen at the medial, anterior and posterior fracture lines. the patient was advanced to full-weightbearing at 24 weeks post-operatively and eventually returned to her baseline level of functioning.
One patient elected to follow-up locally and no follow-up data was obtained while another withdrew from the study after the 12 week follow-up. This patient had bridging callus identified both medially and anteriorly at 6 weeks, but no callus was identified posteriorly. Two patients failed to heal and both underwent revision surgery. One patient required revision with bone graft and additional proximal FCL screws at 12 weeks after continued pain and radiographic concern for slow progress to union. Plate failure occurred in another patient at 9 months post-operatively, which needed revision surgery with bone graft (Figure 2). Prior to hardware failure, only medial bridging callus was identified at 24 weeks post-operatively. This patient complained of persistent pain following the 12 and 24 week follow-up visits after initiation of progressive weight bearing and advancement of physical therapy.
Figure 2. Pre-operative, 24 week and 9 month post-operative radiographs of a 75 year old female who sustained a closed 33A2 periprosthetic distal femur fracture following a ground-level fall. callus formed medially but no anterior or posterior callus is seen at 24 week post-operative films. At this visit she noted progressive pain with increasing activity. the patient presented at 9 months post-operatively with hardware failure and fracture displacement requiring revision surgery.
Discussion
Locking plates have become one of the most common methods of operative treatment of periprosthetic distal femur fractures above a total knee arthroplasty. However, complications associated with locking plates have remained high. Controversy exists as to whether locked plating inhibits secondary bone healing as it limits interfragmentary motion. In-vivo ovine studies have shown increased callus formation with far cortical locking plates when compared to traditional locked plating, possibly due to increased motion allowed by FCL constructs. However, no direct comparison between FCL constructs and traditional locking constructs have been reported.
A literature review of healing of periprosthetic distal femur fractures treated with traditional locking plates and screws identified nine studies (Table 3). The number of fractures treated, number of open fractures, rates of non-union, delayed union, revision or need for secondary bone grafting, implant failures as well as the average time to radiographic healing was included in the review. If the information was not reported, the column was left blank. The average time to radiographic healing ranged from 10 – 24 weeks. The rate of non-union and need for a revision surgery varied from 0 – 22% and the rate of implant failure was as high as 26%. Unfortunately, in the cited articles it was often impossible to know when patients were included in more than one complication category; for example it was not always clear whether a patient that developed a nonunion was the same patient undergoing revision surgery. This makes accurate calculation of the total complication rate difficult but the results in the table suggest there are difficulties with healing and fixation of these fractures.
Table 3.
Healing of Distal Femur Periprosthetic Fractures Treated With Locking Plates
| Author, Year | Fractures (n) | Open (%) | Non Union(%) | Delayed Union(%) | Bone Graft or Implant Revision (%) | Implant Failures (%) | Avg time to heal (wk) |
|---|---|---|---|---|---|---|---|
| Hou et al, 2012 | 34 | 8.8 | 8.8 | 8.8 | 12 | ||
| Large et al,2008 | 25 | 0 | 0 | 4.0 | 0 | ||
| Ricci et al, 2006 | 22 | 0 | 13.6 | 0 | 18.2 | 12 | |
| Hoffman et al, 2012 | 36 | 5.6 | 22.2 | 13.8 | 8.3 | ||
| Ebraheim et al, 2012 | 27 | 3.7 | 7.4 | 26 | 26 | 18 | |
| Streubel et al, 2010 | 61 | 4.9 | 13.1 | 11.5 | 9.8 | 11.5 | |
| Kolb et al, 2010 | 23 | 0 | 0 | 8.7 | 4.3 | 4.3 | 14 |
| Fulkerson et al, 2007 | 18 | 11.1 | 11.1 | 22.2 | 5.6 | 24 | |
| Ehlinger et al, 2011 | 16 | 0 | 6.3 | 0 | 0 | 10 |
The current study found a healing rate of 89%. The two patients that did not heal complained of persistent pain with ambulation and radiographs failed to show evidence of bridging callus. These initial results are comparable to results reported for distal femur periprosthetic fractures treated with locking plate fixation without FCL screws. Although we tried, we found it impossible to compare time to healing between studies given the wide variability in how this data is reported and the small patient numbers. It has been the impression of the authors that callus appears earlier and is more robust and uniform in all three cortices in FCL cases compared to their previous experiences with traditional locking plates for periprosthetic distal femur fractures. Additionally, we did not note any hardware complications related to the FCL screws themselves. Our lone hardware failure involved only the plate at 9 months post-operatively.
The small number of patients in the current study is a significant limitation to assessing differences in healing rates between traditional locking plates and FCL locking plates. Further patient enrollment and follow- up is necessary to adequately compare periprosthetic FCL constructs and traditional locking plate constructs. There are several additional limitations to the current study. Our method of assessing bone healing by assessing callus along the three visible fracture surfaces may not accurately determine whether or not the fracture is healed. Larger studies with longer follow up will need to employ existing validated callus extraction algorithms for objective assessment of periosteal callus formation or 3D imaging using CT. Also, patient clinical data was not included in this study. Further studies will need to combine radiographic outcomes with clinical outcomes including pain, weight-bearing status and return to pre-injury function. Lastly, accurate complication rates in the previously published literature were difficult to determine as the majority of articles did not clarify whether or not the same patients were included in multiple complication categories. This significantly limits the ability to compare these studies to current results.
Future studies will require increased patient enrollment with longer follow-up periods, objective assessment of callus formation and correlation with clinical outcomes. If this data suggests an advantage of FCL screws, a direct comparison of FCL constructs to traditional locking plate constructs in a prospective study is warranted. This study points out the difficulties of proving whether one implant leads to better healing than another in a clinical study with many uncontrolled variables.
References
- 1.Streubel P, Gardner M, Morshed S, Collinge C, Gallagher B, Ricci W. Are extreme distal periprosthetic supracondylar fractures of the femur too distal to fix using a lateral locked plate? J Bone Joint Surg Br. 2010;92-B:527–34. doi: 10.1302/0301-620X.92B3.22996. [DOI] [PubMed] [Google Scholar]
- 2.Healy w, siliski J, Incavo s. Operative treatment of distal femoral fractures proximal to total knee replacements. J Bone Joint Surg Am. 1993;75:27–34. doi: 10.2106/00004623-199301000-00005. [DOI] [PubMed] [Google Scholar]
- 3.Henderson c, Kuhl L, Fitzpatrick D, Marsh J. Locking plates for distal femur fractures: Is there a problem with fracture healing? J Orthop Trauma. 2011;25:S8–S14. doi: 10.1097/BOT.0b013e3182070127. [DOI] [PubMed] [Google Scholar]
- 4.Bottlang M, Feist F. Biomechanics of far cortical locking. J Orthop Trauma. 2011;25(suppl 1):S21–S28. doi: 10.1097/BOT.0b013e318207885b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bottlang M, Lesser M, Koerber J, Doornink J, von Rechenberg B, Augat P, Fitzpatrick D, Madey S, Marsh J. Far cortical locking can improve healing of fractures stabilized with locking plates. J Bone Joint Surg Am. 2010;92:1652–60. doi: 10.2106/JBJS.I.01111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.whelan D, Bhandari M, stephen D, Kreder H, McKee M, Zdero R, schemitsch E. Development of the radiographic union score for tibial fractures for the assessment of tibial fracture healing after intramedullary fixation. J Trauma. 2010;68:629–632. doi: 10.1097/TA.0b013e3181a7c16d. [DOI] [PubMed] [Google Scholar]
- 7.Hou Z, Bowen T, Irgit K, strohecker K, Matzko M, Widmaier J, smith W. Locked plating of periprosthetic femur fractures above total knee arthroplasty. J Orthop Trauma. 2012;26:427–432. doi: 10.1097/BOT.0b013e31822c050b. [DOI] [PubMed] [Google Scholar]
- 8.Large T, Kellam J, Bosse M, sims S, Althausen P, Masonis J. Locked plating of supracondylar periprosthetic femur fractures. J Arthroplasty. 2008;23:S115–120. doi: 10.1016/j.arth.2008.04.021. [DOI] [PubMed] [Google Scholar]
- 9.Hoffmann M, Jones C, sietsema D, Koenig S, Tornetta P. Outcome of periprosthetic distal femoral fractures following knee arthroplasty. Injury. 2012;43:1084–1089. doi: 10.1016/j.injury.2012.01.025. [DOI] [PubMed] [Google Scholar]
- 10.Ebraheim N, Liu J, Hashmi S, sochacki K, Moral M, Hirschfeld A. High complication rate in locking plate fixation of lower periprosthetic distal femur fractures in patients with total knee arthroplasties. J Arthroplasty. 2012;27:809–813. doi: 10.1016/j.arth.2011.08.007. [DOI] [PubMed] [Google Scholar]
- 11.Ricci W, Loftus T, Cox C, Borrelli J. Locked plates combined with minimally invasive insertion technique for the treatment of periprosthetic supracondylar femur fractures above a total knee arthroplasty. J Orthop Trauma. 2006;20:190–196. doi: 10.1097/00005131-200603000-00005. [DOI] [PubMed] [Google Scholar]
- 12.Kolb W, Guhlmann H, Windisch C, Marx F, Koller H, Kolb K. Fixation of periprosthetic femur fractures above total knee arthroplasty with less invasive stabilization system: a midterm follow-up study. J Trauma. 2010;69:670–676. doi: 10.1097/TA.0b013e3181c9ba3b. [DOI] [PubMed] [Google Scholar]
- 13.Ehlinger M, Philippe A, Abane L, Rahme M, Moor B, Arlettaz Y, Bonnemet F. Treatment of periprosthetic femoral fractures of the knee. Knee Surg Sports Traumatol Arthrosc. 2011;19:1473–1478. doi: 10.1007/s00167-011-1480-6. [DOI] [PubMed] [Google Scholar]
- 14.Fulkerson E, Tejwani N, stuchin S, Egol K. Management of periprosthetic femur fractures with a first generation locking plate. Injury. 2007;38:965–972. doi: 10.1016/j.injury.2007.02.026. [DOI] [PubMed] [Google Scholar]