The Journal of Arthroplasty xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthroplasty R. Michael Meneghini, MD a, Brian J. Keyes, DO a, Kartheek K. Reddy, MD a, Dean C. Maar, MD b a b
Department of Orthopaedic Surgery Indiana University School of Medicine, Indianapolis, Indiana Indiana Orthopaedic Hospital–OrthoIndy, Indianapolis, Indiana
a r t i c l e
i n f o
Article history: Received 11 January 2014 Accepted 22 January 2014 Available online xxxx Keywords: periprosthetic fracture total knee arthroplasty knee replacement intramedullary nail locked plate fracture union
a b s t r a c t This study purpose is to analyze outcomes of modern intramedullary (IM) nails with a locked distal screw versus periarticular locking plates for peri-prosthetic supracondylar femur fractures in TKA. Ninety-five consecutive fractures in 91 patients were retrospectively reviewed. Fixation included 29 knees with a retrograde IM nail and 66 periarticular locked plates. Six patients died and 4 were lost to follow-up. There were 2 (9%) nonunions in the IM nail group and 12 non-unions/delayed-unions (19%) in the locked plate group (P = 0.34). A mean of 5.0 distal screws was used in locked plates versus 3.8 distal screws in the IM nails (P b 0.001). Despite a greater quantity of screws in the distal fragment, the failure rate of locked plating was twice that of IM nail fixation. © 2014 Elsevier Inc. All rights reserved.
Distal periprosthetic femur fractures above total knee arthroplasties are difficult fractures to treat for orthopaedic surgeons. With increasing numbers of total knee arthroplasties performed, this specific complication is becoming more prevalent. The management of these fractures is complicated by several variables, which include osteoporotic bone in the distal femoral metaphyseal region, short distal segments for adequate fixation, surgical exposure and blood loss, nonunion, malunion, and malalignment. Recently, peri-articular locked plating techniques have become popular as the most common fixation method. However, consensus does not currently exist within the orthopaedic community or the peer-reviewed literature to establish which implant, fixation strategy or technique is optimal for a successful treatment outcome. The purpose of this study is to retrospectively compare the results of modern IM nails with locked distal fixation to periarticular locked plating. Our hypothesis is that the modern nails will have equivalent success to the periarticular locked plates in achieving clinical and radiographic union. Materials and Methods Institutional review board approval was obtained from three participating centers. Ninety-five consecutive fractures in 91 patients
The Conflict of Interest statement associated with this article can be found at http:// dx.doi.org/10.1016/j.arth.2014.01.025. Reprint requests: R. Michael Meneghini, MD, Department of Orthopaedic Surgery Indiana University School of Medicine 13100 136th Street, Suite 2000 Fishers, IN 46037.
were identified from the institutional database, between 2001 and 2011, as having undergone fixation of a periprosthetic supracondylar femur fracture. There were 10 orthopedic surgeons that participated in the operative management of the study cohort. Fractures were classified using the Lewis and Rorabeck system and subclassified using the AO/OTA classification system. All fractures in this study were Lewis and Rorabeck type II: displaced fractures with a stable prosthesis [1]. Fixation was performed in 29 knees with a retrograde IM nail with a locked distal screw and in 66 periarticular locked plates. Patient demographic, intraoperative, and postoperative data were prospectively collected and recorded from the institutional database. Preoperative variables that were retrieved included age, gender, surgical date, femoral component type, ambulatory level, and fracture classification. In addition, medical comorbidities that may affect fracture healing were documented and included diabetes, smoking, thyroid disease, autoimmune disease and osteoporosis. Postoperative variables included clinical outcome, operative time, time to union, time to weight bearing, ambulatory level, as well as the total number of screws used and number of screws used in the distal fragment. Radiographic analysis was completed using standard AP and lateral knee radiographs. The institutional digital PACS system was utilized for radiographic interpretation. The digital goniometer provided by the PACS software was utilized to analyze pre-operative and post-operative radiographs for angular and translational measurements. The initial postoperative film was analyzed followed by the latest follow-up images available. The highest quality sagittal and coronal images from each time point were used in the radiographic analysis [2,3]. The individual femoral component coronal alignment was measured and recorded from the AP films. The femoral
0883-5403/0000-0000$36.00/0 – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.arth.2014.01.025
Please cite this article as: Meneghini RM, et al, Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthr..., J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.01.025
2
R.M. Meneghini et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
component sagittal alignment was measured and recorded from the lateral radiographic projections. For the analysis of the coronal plane the distal femoral valgus angle and translation of the distal femoral fragment were assessed. Selecting a midpoint of the femoral component and the femoral diaphysis, then drawing two parallel lines tangential to the anatomic axis, and measuring the translation distance between these two lines determined the translation measurement (Fig. 1). Clinical and radiographic follow up was obtained until fracture union. Nonunion or delayed union was considered a clinical failure. Multivariate statistical analysis was performed to assess differences in outcomes between groups and variables predictive of nonunion or failure. Differences were considered statistically significant at P b 0.05. Results Six patients died and 4 patients were lost to follow-up, leaving 85 TKR periprosthetic fractures (22 IM nails, 63 locked plates) with a minimum of 6 weeks follow-up (mean 31.8 weeks, range 6 to 176 weeks) (Table 1). Twenty-five of original 29 fractures in the IM nail group were in knees with cruciate-retaining implant designs and the other four (13.8%) were in posterior-substituting designs with an open box amendable to utilization of an IM nail. Forty-six of 66 original fractures in the locked-plate group were in knees with cruciate-retaining implant designs and the other 20 were in posteriorsubstituting designs (30.3%). Seventy-one of 85 knees (83.5%) went on to union at an average of 16 weeks. There were 2 (9%) nonunions in the IM nail group and 12 nonunions or delayed unions (19%) in the
locked plate group (P = 0.34) (Table 2). Eleven of the 12 non-unions/ delayed-unions in the locked plate group were in females, and one of the two IM nail group non-unions was a female. There was no difference in time to union between groups (P = 0.64). A mean of 5.0 (range, 3–8) distal screws were used in the locked plate group, compared to 3.8 (range, 3–4) distal screws in the IM nail group (P b 0.001). Radiographic analysis (Table 3) demonstrated the mean immediate postoperative alignment of the femoral component in the sagittal plane was 0.36°of extension in the locking plate group, which was not statistically different than the 1.05° extension in the IM nail group (P = 0.7). At final radiographic follow up, the sagittal femoral flexion changed minimally to mean of 0.2° in the locking plates and 0.26° in the IM nails (P = 0.9). The immediate postoperative femoral component anatomic valgus in the coronal plane was measured at a mean of 5.62° valgus in the locking plate group, compared to 6.4° in the IM nail group (P = 0.7). At final radiographic follow-up, anatomic coronal alignment changed minimally to a mean of 5.28° valgus in the locked plate group, compared to 6.2° in the IM nail group (P = 0.7). Further, “acceptable” coronal femoral valgus alignment was defined as 5˚ ± 2° anatomic femoral valgus, which revealed 56% of IM nails and 41% of locking plates fell outside of acceptable alignment at final radiographic follow-up. As might be expected with these devices, the locking plate cohort demonstrated greater immediate postoperative translation (distal fragment translation medially) of 4.7 mm compared to 1.8 mm in the IM nail group (P = 0.04) (Table 3); however, at final follow-up there was no statistically significant difference between the study groups in regard to distal femoral fragment translation with means of 5.3 mm translation in the locking plates and 2.4 mm in the IM nail group (P = 0.1). Surprisingly, none of the medical comorbidities were statistically correlated or predictive of failure or nonunion. With respect to ambulatory status, the IM nail group demonstrated an overall lower ambulation level (as indicated by the percentage of patients who use a walking aid) preoperatively compared to the locking plate group (P = 0.013), indicating a potential selection bias of the surgeons to use the IM nails in more sedentary and less mobile patients. The mean time to resume full weight-bearing was less in the IM nail group at 9.1 weeks compared to 11.7 weeks in the locked-plate group (P = 0.01). Discussion
Fig. 1. AP radiograph of healed distal femoral periprosthetic femur fracture above a well-fixed TKA demonstrating the method for measuring distal fragment translation.
Supracondylar periprosthetic distal femur fractures after total knee arthroplasty (TKA) are problematic for the treating orthopedic surgeon. These fractures represent a rare but challenging problem, affecting 0.3% to 2.5% of total knee arthroplasty patients [4,5]. The treatment challenges arise from the typical distal fracture location and poor bone quality, the technical challenges associated with the femoral implant and varying designs, the less than optimal bone biology and the elderly patient population in which these fractures typically occur. It is well reported that these fractures have not responded well to traditional non-locked plating techniques. Moran et al [6] reported satisfactory results in 10 of 15 patients treated with non-locked plates. The authors noted that although better outcomes resulted from operative vs. non-operative treatment for displaced fractures, the complication rate of non-locked plating techniques were substantial. The results of internal fixation of supracondylar periprosthetic distal femur fractures have been documented, and have almost universally defined union and alignment as their outcome measures. Satisfactory outcomes have been reported with use of both periarticular locked plates and retrograde intramedullary (IM) nail fixation. Ricci et al [7] utilized minimally invasive locked plating techniques in periprosthetic fractures, of which 19 of 22 fractures united with mean follow-up of 15 months. When examining the Less Invasive
Please cite this article as: Meneghini RM, et al, Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthr..., J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.01.025
R.M. Meneghini et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
3
Table 1 Preoperative Variables.
Age Gender Comorbidities
Women Diabetes Smoking Renal Neuro-muscular Thyroid Obesity RA b= 2005 2006–2008 2009–2011 None Cane Walker Nonambulator A C 1 2 3
Year
Ambulatory devices
Fracture Type Syb type
Nail
Plate
Total
N = 19
N = 62
N = 81
N
(%)
N
(%)
N
(%)
74 14 10 2 3 2 1 1 1 6 5 8 7 3 6 3 18 1 6 3 10
(67–81) (74) (53) (11) (16) (11) (5) (5) (5) (32) (26) (42) (37) (16) (32) (16) (95) (5) (32) (16) (53)
74 51 19 2 8 0 5 1 4 9 27 26 33 12 17 0 61 1 22 6 34
(62–82) (82) (31) (3) (13) (67) (8) (2) (6) (15) (44) (42) (53) (19) (27) (0) (98) (2) (35) (10) (55)
74 65 29 4 11 2 6 2 5 15 32 34 40 15 23 3 83 2 28 9 44
(64–81) (80) (36) (5) (14) (2) (7) (2) (6) (19) (40) (42) (49) (19) (28) (4) (102) (2) (35) (11) (54)
P Value 1 0.51 0.10 0.23 0.71 0.05 1 0.42 1 0.18
0.01
0.42 0.75
P values: Fisher’s exact test for two category variables; Chi-Square for variables greater than two categories.
Stabilization System (LISS) locked plating technique, Fulkerson et al [8] demonstrated that 21 of 24 fractures united at mean follow-up of 6 months. These retrospective reports of relatively small cohorts have demonstrated acceptable union rates with locked plate fixation, which have led to relatively widespread adoption of this fixation technique for these difficult fractures. With regard to retrograde IM nailing techniques, excellent union rates have been reported as well. Gliatis et al [9], Han et al [10], and Chettiar et al [11] examined a total of 32 periprosthetic supracondylar distal femur fractures treated with retrograde IM nailing and collectively reported a 100% union rate, with only 1 reported fracture healing in malalignment. However, the cohort sizes in these studies are small and non-comparative, which fail to delineate a superior fixation strategy for the treatment of these complex fractures. In addition, a lack of data exists to direct the surgeon toward a fixation strategy that might be optimal for a given patient type, bone morphology or clinical scenario. There is also a paucity of data comparing modern intramedullary nails with fixed angle interlocking screws to periarticular locking plates. The purported advantages of the modern nails lie in their unique design and ability to transform into a fixed angle device. These designs incorporate fixed angle interlocking screws that lie distal to
the anterior flange of the prosthesis, and subsequently are positioned to allow enhanced fixation in the distal fracture fragment. To the authors’ knowledge there are only five studies that have compared retrograde IM nailing to periarticular locked plating for the treatment of these fractures. Large et al [12] compared union rates between 29 locked plates versus a group of 21 non-locked plates/ retrograde IM nails. The study conclusion was that locked plating outperformed retrograde IM nailing and non-locked plating techniques. Muller et al [13] compared treatment of 9 LISS Plates vs 9 Retrograde IM nails. One patient in each group required revision surgery and 1 patient in the retrograde IM group malunited in 18° valgus and the authors noted that the LISS plate was the superior implant for fracture patterns with a small distal fragment. Althausen et al [14] compared four treatment methods in only 12 patients. The authors concluded that all constructs frequently achieve limited distal fixation, leading to loss of fixation and varus angulation. However, this complication was not observed with the LISS plating group. Recently, Hou et al [15] reported a comparison of 34 periarticular locked plates to 18 retrograde nails. In the plate group, there were three (9%) nonunions and three (9%) malunions. In the nail group, one (6%) failed to unite as a result of infection and two (11%) developed fracture malunions. The authors commented that their results did not
Table 2 Demographics and Follow-Up of Non-Unions and Delayed-Unions. Patient 1 2 3 4 5 6 7 8 9 10 11 12 1 2
Fixation Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate IM Nail IM Nail
Gender
Age
F F F M F F F F F F F F M F
81 62 54 50 80 83 65 85 58 86 89 63 66 61
Comorbidities Diabetes
Diabetes
Diabetes, Kidney Disease Breast CA Rheumatoid Arthritis Kidney Disease
Smoker
Time to Union
Classification
60 weeks 52 weeks 51 weeks 28 weeks 32 weeks Non-union Non-union Non-union Non-union Non-union Non-union Non-union Non-union Non-union
33A3.3 33A1.2 33A3.2 33A1.3 33A1.2 33A2.2 33A2.3 33A3.3 33A3.3 33A3.2 33A3.2 33A1.2 32C2.3 33A3.3
Final Follow up
Intervention Bone Stimulator Bone Stimulator Bone stimulator
24 weeks 7 weeks 17 weeks 21 weeks 38 weeks 45 weeks 57 weeks 39 weeks 36 weeks
Revision Locked-Plate with bone grafting Converted to distal femoral arthroplasty Revision Locked-Plate with bone grafting Revision Locked-Plate with bone grafting Converted to distal femoral arthroplasty Revision Locked-Plate with bone grafting Bone Stimulator Revision to Locked-Plate
Please cite this article as: Meneghini RM, et al, Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthr..., J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.01.025
4
R.M. Meneghini et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
Table 3 Radiographic Mean Values.
Immediate Postop Femoral Flexion Final Postop Femoral Flexion Immediate Postop Femoral Valgus Final Postop Femoral Valgus Immediate Postop Femoral Translation Final Postop Femoral Translation
Plate
Nail
0.36o ext 0.20o flex 5.62o 5.28o 4.7 mm 5.3 mm
1.05o ext 0.26o flex 6.4o 6.2o 1.8 mm 2.4 mm
P Value 0.7 0.9 0.7 0.7 0.04 0.1
meet statistical significance due to the small cohort size. It is apparent that due to the small cohort sizes of the all above studies, no definitive conclusions can be made regarding the optimal fixation method for these peri-prosthetic fractures around TKA. Furthermore, none of the above comparative studies analyzed the modern IM nails with fixed angle distal fixation capabilities. The strengths of this study lie in the comparative cohorts of fixation type, the analysis of the unique and modern IM nail with fixed angle capability and the relatively large number of patients in the study. To our knowledge, this is the largest consecutive series in the literature documenting operative fixation of supracondylar periprosthetic femur fractures after TKA. The study results and union rates are comparable to those discussed above, with a 9% failure rate in the IM nail group and a 19% failure rate in the locked plate group. The documented nonunion rate of this study, as well as others in the literature, underscores the difficulty in treating these fractures. Additionally, the findings of this study reveal that modern retrograde IM nail fixation represents a viable technique for treatment of periprosthetic femur fractures that is at least equivalent to the more popularized locking plates. In fact, despite having a greater quantity of screws in the distal fragment and having a higher preoperative ambulation level, the failure rate of locked plate fixation in this study is twice that of IM nail fixation. The purported advantages of the modern nails lie in the preservation of the local fracture biology, which may facilitate healing and subsequent union, as well as the modern implant design. The modern designs incorporate fixed angle interlocking screws that lie distal to the anterior flange of the prosthesis. This added distal fixation allows multiple fixed angle fixation points in the often very short distal segment. Supporting the mechanical integrity and feasibility of this fixation method, Bong et al [16] performed a cadaveric study, which showed retrograde IM nailing was biomechanically superior to locked plating in management of periprosthetic distal femur fractures. The limitations of this study include the retrospective nature of the analysis, the smaller quantity of patients in the IM nail group, the short clinical follow up and the apparent surgeon bias for selecting the IM nail fixation in patients of a lower ambulatory status. An additional selection bias may exist based on the type of TKA implant, with a greater percentage of patients in the locked-plate group having a PS design due to the technical challenge of using an IM nail in a TKA with a closed-box or box of inadequate dimension to accommodate the IM nail. However, despite having a smaller number of patients in the IM nail cohort, the 22 patients represent one of the largest studies of this fixation to date. In addition, it is the only study to our knowledge reporting the fixed angle modern IM nail construct. The short clinical follow up does not allow long term functional outcome to be
measured and is therefore limited to information on union and early failure rates. It would be useful to have the longer-term follow up and further analysis of the functional status of the patients and TKA implant survivorship over time. Finally, there was an apparent bias in the study, as the IM nail group demonstrated an overall lower ambulation level preoperatively compared to the locking plate group. However, it is reasonable to conclude the study findings of equivalent, or potentially superior, union rates found in the IM nail group remain valid despite the selection bias. Less ambulatory patients would be expected to have less optimal bone quality due to the sedentary activity levels in those patients, yet despite that assumption, the IM nail group demonstrated a failure rate 50% less than the locked periarticular plate group. In summary, further research and clinical follow-up are needed to ascertain if one type of fixation method is superior to another in the challenging treatment of periprosthetic distal femur fractures after TKR. However, the results of this study support the use of modern IM nail fixation with locked distal screw fixation when compared to locking periarticular plate fixation. Despite the popularity of locked plates, the authors encourage other arthroplasty and trauma surgeons who treat these challenging problems to consider the modern retrograde IM nail in their armamentarium of treatment strategies for these periprosthetic fractures. References 1. Lewis PL, Rorabeck CH, Angliss RD. Fractures of the femur, tibia, and patella after total knee arthroplasty: decision making and principles of management. Instr Course Lect 1998;47:449. 2. Ewald FC. The knee society total knee arthroplasty roentgenographic evaluation and scoring system. Clin Orthop 1989;24:9. 3. Bargren JH, Blaha MD, Freeman MAR. Alignment in total knee arthroplasty: correlated biomechanical and clinical observations. Clin Orthop 1983;173:178. 4. Delport PH, Van Auekercke R, Martens M, et al. Conservative treatment of ipsilateral supracondylar femoral fracture after total knee arthroplasty. J Trauma 1984;24:846. 5. Merkel KD, Johnson EW. Supracondylar fracture of the knee after total knee arthroplasty. J Bone Joint Surg Am 1986;68:29. 6. Moran MC, Brick GW, Sledge CB, et al. Supracondylar femoral fracture following total. 7. Ricci WM, Loftus T, Cox C, et al. 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(3):190. 8. Fulkerson E, Tejwani N, Stuchin S, et al. Management of periprosthetic femur fractures with a first generation locking plate. Injury 2007;38(8):965. 9. Gliatis J, Megas P, Panagiotopoulos E, et al. Midterm results of treatment with a retrograde nail for supracondylar periprosthetic fractures of the femur following total knee arthroplasty. J Orthop Trauma 2005;19(3):164. 10. Han HS, Oh KW, Kang SB. Retrograde Intramedullary nailing for periprosthetic supracondylar fractures of the femur after total knee arthroplasty. Clin Orthop Surg 2009;1(4):201. 11. Chettiar K, Jackson MP, Brewin J, et al. Supracondylar periprosthetic femoral fractures following total knee arthroplasty: treatment with a retrograde intramedullary nail. Int Orthop 2009;33(4):981. 12. Large TM, Kellam JF, Bosse MJ, et al. Locked plating of supracondylar periprosthetic femur fractures. J Arthroplasty 2008;23(Suppl 1):115. 13. Muller EJ, Kutscha-Lissberg F, et al. Periprosthetic supracondylar femoral fractures: LISS or retrograde intramedullary nailing? Problems with the use of minimally invasive technique [in German]. Unfallchirurg 2004;107:181. 14. Althausen PL, Lee MA, Finkemeier CG, et al. Operative stabilization of supracondylar femur fractures above total knee arthroplasty: a comparison of four treatment methods. J Arthroplasty 2003;18(7):834. 15. Hou Z, Bowen TR, Irgit K, et al. Locked plating of periprosthetic femur fractures above total knee arthroplasty. J Orthop Trauma 2012;26(7):427. 16. Bong MR, Egol KA, Koval KJ, et al. Comparison of the LISS and a retrograde-inserted supracondylar intramedullary nail for fixation of a periprosthetic distal femur fracture proximal to a total knee arthroplasty. J Arthroplasty 2002;17:876.
Please cite this article as: Meneghini RM, et al, Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthr..., J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.01.025
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthroplasty R. Michael Meneghini, MD a, Brian J. Keyes, DO a, Kartheek K. Reddy, MD a, Dean C. Maar, MD b a b
Department of Orthopaedic Surgery Indiana University School of Medicine, Indianapolis, Indiana Indiana Orthopaedic Hospital–OrthoIndy, Indianapolis, Indiana
a r t i c l e
i n f o
Article history: Received 11 January 2014 Accepted 22 January 2014 Available online xxxx Keywords: periprosthetic fracture total knee arthroplasty knee replacement intramedullary nail locked plate fracture union
a b s t r a c t This study purpose is to analyze outcomes of modern intramedullary (IM) nails with a locked distal screw versus periarticular locking plates for peri-prosthetic supracondylar femur fractures in TKA. Ninety-five consecutive fractures in 91 patients were retrospectively reviewed. Fixation included 29 knees with a retrograde IM nail and 66 periarticular locked plates. Six patients died and 4 were lost to follow-up. There were 2 (9%) nonunions in the IM nail group and 12 non-unions/delayed-unions (19%) in the locked plate group (P = 0.34). A mean of 5.0 distal screws was used in locked plates versus 3.8 distal screws in the IM nails (P b 0.001). Despite a greater quantity of screws in the distal fragment, the failure rate of locked plating was twice that of IM nail fixation. © 2014 Elsevier Inc. All rights reserved.
Distal periprosthetic femur fractures above total knee arthroplasties are difficult fractures to treat for orthopaedic surgeons. With increasing numbers of total knee arthroplasties performed, this specific complication is becoming more prevalent. The management of these fractures is complicated by several variables, which include osteoporotic bone in the distal femoral metaphyseal region, short distal segments for adequate fixation, surgical exposure and blood loss, nonunion, malunion, and malalignment. Recently, peri-articular locked plating techniques have become popular as the most common fixation method. However, consensus does not currently exist within the orthopaedic community or the peer-reviewed literature to establish which implant, fixation strategy or technique is optimal for a successful treatment outcome. The purpose of this study is to retrospectively compare the results of modern IM nails with locked distal fixation to periarticular locked plating. Our hypothesis is that the modern nails will have equivalent success to the periarticular locked plates in achieving clinical and radiographic union. Materials and Methods Institutional review board approval was obtained from three participating centers. Ninety-five consecutive fractures in 91 patients
The Conflict of Interest statement associated with this article can be found at http:// dx.doi.org/10.1016/j.arth.2014.01.025. Reprint requests: R. Michael Meneghini, MD, Department of Orthopaedic Surgery Indiana University School of Medicine 13100 136th Street, Suite 2000 Fishers, IN 46037.
were identified from the institutional database, between 2001 and 2011, as having undergone fixation of a periprosthetic supracondylar femur fracture. There were 10 orthopedic surgeons that participated in the operative management of the study cohort. Fractures were classified using the Lewis and Rorabeck system and subclassified using the AO/OTA classification system. All fractures in this study were Lewis and Rorabeck type II: displaced fractures with a stable prosthesis [1]. Fixation was performed in 29 knees with a retrograde IM nail with a locked distal screw and in 66 periarticular locked plates. Patient demographic, intraoperative, and postoperative data were prospectively collected and recorded from the institutional database. Preoperative variables that were retrieved included age, gender, surgical date, femoral component type, ambulatory level, and fracture classification. In addition, medical comorbidities that may affect fracture healing were documented and included diabetes, smoking, thyroid disease, autoimmune disease and osteoporosis. Postoperative variables included clinical outcome, operative time, time to union, time to weight bearing, ambulatory level, as well as the total number of screws used and number of screws used in the distal fragment. Radiographic analysis was completed using standard AP and lateral knee radiographs. The institutional digital PACS system was utilized for radiographic interpretation. The digital goniometer provided by the PACS software was utilized to analyze pre-operative and post-operative radiographs for angular and translational measurements. The initial postoperative film was analyzed followed by the latest follow-up images available. The highest quality sagittal and coronal images from each time point were used in the radiographic analysis [2,3]. The individual femoral component coronal alignment was measured and recorded from the AP films. The femoral
0883-5403/0000-0000$36.00/0 – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.arth.2014.01.025
Please cite this article as: Meneghini RM, et al, Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthr..., J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.01.025
2
R.M. Meneghini et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
component sagittal alignment was measured and recorded from the lateral radiographic projections. For the analysis of the coronal plane the distal femoral valgus angle and translation of the distal femoral fragment were assessed. Selecting a midpoint of the femoral component and the femoral diaphysis, then drawing two parallel lines tangential to the anatomic axis, and measuring the translation distance between these two lines determined the translation measurement (Fig. 1). Clinical and radiographic follow up was obtained until fracture union. Nonunion or delayed union was considered a clinical failure. Multivariate statistical analysis was performed to assess differences in outcomes between groups and variables predictive of nonunion or failure. Differences were considered statistically significant at P b 0.05. Results Six patients died and 4 patients were lost to follow-up, leaving 85 TKR periprosthetic fractures (22 IM nails, 63 locked plates) with a minimum of 6 weeks follow-up (mean 31.8 weeks, range 6 to 176 weeks) (Table 1). Twenty-five of original 29 fractures in the IM nail group were in knees with cruciate-retaining implant designs and the other four (13.8%) were in posterior-substituting designs with an open box amendable to utilization of an IM nail. Forty-six of 66 original fractures in the locked-plate group were in knees with cruciate-retaining implant designs and the other 20 were in posteriorsubstituting designs (30.3%). Seventy-one of 85 knees (83.5%) went on to union at an average of 16 weeks. There were 2 (9%) nonunions in the IM nail group and 12 nonunions or delayed unions (19%) in the
locked plate group (P = 0.34) (Table 2). Eleven of the 12 non-unions/ delayed-unions in the locked plate group were in females, and one of the two IM nail group non-unions was a female. There was no difference in time to union between groups (P = 0.64). A mean of 5.0 (range, 3–8) distal screws were used in the locked plate group, compared to 3.8 (range, 3–4) distal screws in the IM nail group (P b 0.001). Radiographic analysis (Table 3) demonstrated the mean immediate postoperative alignment of the femoral component in the sagittal plane was 0.36°of extension in the locking plate group, which was not statistically different than the 1.05° extension in the IM nail group (P = 0.7). At final radiographic follow up, the sagittal femoral flexion changed minimally to mean of 0.2° in the locking plates and 0.26° in the IM nails (P = 0.9). The immediate postoperative femoral component anatomic valgus in the coronal plane was measured at a mean of 5.62° valgus in the locking plate group, compared to 6.4° in the IM nail group (P = 0.7). At final radiographic follow-up, anatomic coronal alignment changed minimally to a mean of 5.28° valgus in the locked plate group, compared to 6.2° in the IM nail group (P = 0.7). Further, “acceptable” coronal femoral valgus alignment was defined as 5˚ ± 2° anatomic femoral valgus, which revealed 56% of IM nails and 41% of locking plates fell outside of acceptable alignment at final radiographic follow-up. As might be expected with these devices, the locking plate cohort demonstrated greater immediate postoperative translation (distal fragment translation medially) of 4.7 mm compared to 1.8 mm in the IM nail group (P = 0.04) (Table 3); however, at final follow-up there was no statistically significant difference between the study groups in regard to distal femoral fragment translation with means of 5.3 mm translation in the locking plates and 2.4 mm in the IM nail group (P = 0.1). Surprisingly, none of the medical comorbidities were statistically correlated or predictive of failure or nonunion. With respect to ambulatory status, the IM nail group demonstrated an overall lower ambulation level (as indicated by the percentage of patients who use a walking aid) preoperatively compared to the locking plate group (P = 0.013), indicating a potential selection bias of the surgeons to use the IM nails in more sedentary and less mobile patients. The mean time to resume full weight-bearing was less in the IM nail group at 9.1 weeks compared to 11.7 weeks in the locked-plate group (P = 0.01). Discussion
Fig. 1. AP radiograph of healed distal femoral periprosthetic femur fracture above a well-fixed TKA demonstrating the method for measuring distal fragment translation.
Supracondylar periprosthetic distal femur fractures after total knee arthroplasty (TKA) are problematic for the treating orthopedic surgeon. These fractures represent a rare but challenging problem, affecting 0.3% to 2.5% of total knee arthroplasty patients [4,5]. The treatment challenges arise from the typical distal fracture location and poor bone quality, the technical challenges associated with the femoral implant and varying designs, the less than optimal bone biology and the elderly patient population in which these fractures typically occur. It is well reported that these fractures have not responded well to traditional non-locked plating techniques. Moran et al [6] reported satisfactory results in 10 of 15 patients treated with non-locked plates. The authors noted that although better outcomes resulted from operative vs. non-operative treatment for displaced fractures, the complication rate of non-locked plating techniques were substantial. The results of internal fixation of supracondylar periprosthetic distal femur fractures have been documented, and have almost universally defined union and alignment as their outcome measures. Satisfactory outcomes have been reported with use of both periarticular locked plates and retrograde intramedullary (IM) nail fixation. Ricci et al [7] utilized minimally invasive locked plating techniques in periprosthetic fractures, of which 19 of 22 fractures united with mean follow-up of 15 months. When examining the Less Invasive
Please cite this article as: Meneghini RM, et al, Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthr..., J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.01.025
R.M. Meneghini et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
3
Table 1 Preoperative Variables.
Age Gender Comorbidities
Women Diabetes Smoking Renal Neuro-muscular Thyroid Obesity RA b= 2005 2006–2008 2009–2011 None Cane Walker Nonambulator A C 1 2 3
Year
Ambulatory devices
Fracture Type Syb type
Nail
Plate
Total
N = 19
N = 62
N = 81
N
(%)
N
(%)
N
(%)
74 14 10 2 3 2 1 1 1 6 5 8 7 3 6 3 18 1 6 3 10
(67–81) (74) (53) (11) (16) (11) (5) (5) (5) (32) (26) (42) (37) (16) (32) (16) (95) (5) (32) (16) (53)
74 51 19 2 8 0 5 1 4 9 27 26 33 12 17 0 61 1 22 6 34
(62–82) (82) (31) (3) (13) (67) (8) (2) (6) (15) (44) (42) (53) (19) (27) (0) (98) (2) (35) (10) (55)
74 65 29 4 11 2 6 2 5 15 32 34 40 15 23 3 83 2 28 9 44
(64–81) (80) (36) (5) (14) (2) (7) (2) (6) (19) (40) (42) (49) (19) (28) (4) (102) (2) (35) (11) (54)
P Value 1 0.51 0.10 0.23 0.71 0.05 1 0.42 1 0.18
0.01
0.42 0.75
P values: Fisher’s exact test for two category variables; Chi-Square for variables greater than two categories.
Stabilization System (LISS) locked plating technique, Fulkerson et al [8] demonstrated that 21 of 24 fractures united at mean follow-up of 6 months. These retrospective reports of relatively small cohorts have demonstrated acceptable union rates with locked plate fixation, which have led to relatively widespread adoption of this fixation technique for these difficult fractures. With regard to retrograde IM nailing techniques, excellent union rates have been reported as well. Gliatis et al [9], Han et al [10], and Chettiar et al [11] examined a total of 32 periprosthetic supracondylar distal femur fractures treated with retrograde IM nailing and collectively reported a 100% union rate, with only 1 reported fracture healing in malalignment. However, the cohort sizes in these studies are small and non-comparative, which fail to delineate a superior fixation strategy for the treatment of these complex fractures. In addition, a lack of data exists to direct the surgeon toward a fixation strategy that might be optimal for a given patient type, bone morphology or clinical scenario. There is also a paucity of data comparing modern intramedullary nails with fixed angle interlocking screws to periarticular locking plates. The purported advantages of the modern nails lie in their unique design and ability to transform into a fixed angle device. These designs incorporate fixed angle interlocking screws that lie distal to
the anterior flange of the prosthesis, and subsequently are positioned to allow enhanced fixation in the distal fracture fragment. To the authors’ knowledge there are only five studies that have compared retrograde IM nailing to periarticular locked plating for the treatment of these fractures. Large et al [12] compared union rates between 29 locked plates versus a group of 21 non-locked plates/ retrograde IM nails. The study conclusion was that locked plating outperformed retrograde IM nailing and non-locked plating techniques. Muller et al [13] compared treatment of 9 LISS Plates vs 9 Retrograde IM nails. One patient in each group required revision surgery and 1 patient in the retrograde IM group malunited in 18° valgus and the authors noted that the LISS plate was the superior implant for fracture patterns with a small distal fragment. Althausen et al [14] compared four treatment methods in only 12 patients. The authors concluded that all constructs frequently achieve limited distal fixation, leading to loss of fixation and varus angulation. However, this complication was not observed with the LISS plating group. Recently, Hou et al [15] reported a comparison of 34 periarticular locked plates to 18 retrograde nails. In the plate group, there were three (9%) nonunions and three (9%) malunions. In the nail group, one (6%) failed to unite as a result of infection and two (11%) developed fracture malunions. The authors commented that their results did not
Table 2 Demographics and Follow-Up of Non-Unions and Delayed-Unions. Patient 1 2 3 4 5 6 7 8 9 10 11 12 1 2
Fixation Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate Locked-Plate IM Nail IM Nail
Gender
Age
F F F M F F F F F F F F M F
81 62 54 50 80 83 65 85 58 86 89 63 66 61
Comorbidities Diabetes
Diabetes
Diabetes, Kidney Disease Breast CA Rheumatoid Arthritis Kidney Disease
Smoker
Time to Union
Classification
60 weeks 52 weeks 51 weeks 28 weeks 32 weeks Non-union Non-union Non-union Non-union Non-union Non-union Non-union Non-union Non-union
33A3.3 33A1.2 33A3.2 33A1.3 33A1.2 33A2.2 33A2.3 33A3.3 33A3.3 33A3.2 33A3.2 33A1.2 32C2.3 33A3.3
Final Follow up
Intervention Bone Stimulator Bone Stimulator Bone stimulator
24 weeks 7 weeks 17 weeks 21 weeks 38 weeks 45 weeks 57 weeks 39 weeks 36 weeks
Revision Locked-Plate with bone grafting Converted to distal femoral arthroplasty Revision Locked-Plate with bone grafting Revision Locked-Plate with bone grafting Converted to distal femoral arthroplasty Revision Locked-Plate with bone grafting Bone Stimulator Revision to Locked-Plate
Please cite this article as: Meneghini RM, et al, Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthr..., J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.01.025
4
R.M. Meneghini et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
Table 3 Radiographic Mean Values.
Immediate Postop Femoral Flexion Final Postop Femoral Flexion Immediate Postop Femoral Valgus Final Postop Femoral Valgus Immediate Postop Femoral Translation Final Postop Femoral Translation
Plate
Nail
0.36o ext 0.20o flex 5.62o 5.28o 4.7 mm 5.3 mm
1.05o ext 0.26o flex 6.4o 6.2o 1.8 mm 2.4 mm
P Value 0.7 0.9 0.7 0.7 0.04 0.1
meet statistical significance due to the small cohort size. It is apparent that due to the small cohort sizes of the all above studies, no definitive conclusions can be made regarding the optimal fixation method for these peri-prosthetic fractures around TKA. Furthermore, none of the above comparative studies analyzed the modern IM nails with fixed angle distal fixation capabilities. The strengths of this study lie in the comparative cohorts of fixation type, the analysis of the unique and modern IM nail with fixed angle capability and the relatively large number of patients in the study. To our knowledge, this is the largest consecutive series in the literature documenting operative fixation of supracondylar periprosthetic femur fractures after TKA. The study results and union rates are comparable to those discussed above, with a 9% failure rate in the IM nail group and a 19% failure rate in the locked plate group. The documented nonunion rate of this study, as well as others in the literature, underscores the difficulty in treating these fractures. Additionally, the findings of this study reveal that modern retrograde IM nail fixation represents a viable technique for treatment of periprosthetic femur fractures that is at least equivalent to the more popularized locking plates. In fact, despite having a greater quantity of screws in the distal fragment and having a higher preoperative ambulation level, the failure rate of locked plate fixation in this study is twice that of IM nail fixation. The purported advantages of the modern nails lie in the preservation of the local fracture biology, which may facilitate healing and subsequent union, as well as the modern implant design. The modern designs incorporate fixed angle interlocking screws that lie distal to the anterior flange of the prosthesis. This added distal fixation allows multiple fixed angle fixation points in the often very short distal segment. Supporting the mechanical integrity and feasibility of this fixation method, Bong et al [16] performed a cadaveric study, which showed retrograde IM nailing was biomechanically superior to locked plating in management of periprosthetic distal femur fractures. The limitations of this study include the retrospective nature of the analysis, the smaller quantity of patients in the IM nail group, the short clinical follow up and the apparent surgeon bias for selecting the IM nail fixation in patients of a lower ambulatory status. An additional selection bias may exist based on the type of TKA implant, with a greater percentage of patients in the locked-plate group having a PS design due to the technical challenge of using an IM nail in a TKA with a closed-box or box of inadequate dimension to accommodate the IM nail. However, despite having a smaller number of patients in the IM nail cohort, the 22 patients represent one of the largest studies of this fixation to date. In addition, it is the only study to our knowledge reporting the fixed angle modern IM nail construct. The short clinical follow up does not allow long term functional outcome to be
measured and is therefore limited to information on union and early failure rates. It would be useful to have the longer-term follow up and further analysis of the functional status of the patients and TKA implant survivorship over time. Finally, there was an apparent bias in the study, as the IM nail group demonstrated an overall lower ambulation level preoperatively compared to the locking plate group. However, it is reasonable to conclude the study findings of equivalent, or potentially superior, union rates found in the IM nail group remain valid despite the selection bias. Less ambulatory patients would be expected to have less optimal bone quality due to the sedentary activity levels in those patients, yet despite that assumption, the IM nail group demonstrated a failure rate 50% less than the locked periarticular plate group. In summary, further research and clinical follow-up are needed to ascertain if one type of fixation method is superior to another in the challenging treatment of periprosthetic distal femur fractures after TKR. However, the results of this study support the use of modern IM nail fixation with locked distal screw fixation when compared to locking periarticular plate fixation. Despite the popularity of locked plates, the authors encourage other arthroplasty and trauma surgeons who treat these challenging problems to consider the modern retrograde IM nail in their armamentarium of treatment strategies for these periprosthetic fractures. References 1. Lewis PL, Rorabeck CH, Angliss RD. Fractures of the femur, tibia, and patella after total knee arthroplasty: decision making and principles of management. Instr Course Lect 1998;47:449. 2. Ewald FC. The knee society total knee arthroplasty roentgenographic evaluation and scoring system. Clin Orthop 1989;24:9. 3. Bargren JH, Blaha MD, Freeman MAR. Alignment in total knee arthroplasty: correlated biomechanical and clinical observations. Clin Orthop 1983;173:178. 4. Delport PH, Van Auekercke R, Martens M, et al. Conservative treatment of ipsilateral supracondylar femoral fracture after total knee arthroplasty. J Trauma 1984;24:846. 5. Merkel KD, Johnson EW. Supracondylar fracture of the knee after total knee arthroplasty. J Bone Joint Surg Am 1986;68:29. 6. Moran MC, Brick GW, Sledge CB, et al. Supracondylar femoral fracture following total. 7. Ricci WM, Loftus T, Cox C, et al. 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(3):190. 8. Fulkerson E, Tejwani N, Stuchin S, et al. Management of periprosthetic femur fractures with a first generation locking plate. Injury 2007;38(8):965. 9. Gliatis J, Megas P, Panagiotopoulos E, et al. Midterm results of treatment with a retrograde nail for supracondylar periprosthetic fractures of the femur following total knee arthroplasty. J Orthop Trauma 2005;19(3):164. 10. Han HS, Oh KW, Kang SB. Retrograde Intramedullary nailing for periprosthetic supracondylar fractures of the femur after total knee arthroplasty. Clin Orthop Surg 2009;1(4):201. 11. Chettiar K, Jackson MP, Brewin J, et al. Supracondylar periprosthetic femoral fractures following total knee arthroplasty: treatment with a retrograde intramedullary nail. Int Orthop 2009;33(4):981. 12. Large TM, Kellam JF, Bosse MJ, et al. Locked plating of supracondylar periprosthetic femur fractures. J Arthroplasty 2008;23(Suppl 1):115. 13. Muller EJ, Kutscha-Lissberg F, et al. Periprosthetic supracondylar femoral fractures: LISS or retrograde intramedullary nailing? Problems with the use of minimally invasive technique [in German]. Unfallchirurg 2004;107:181. 14. Althausen PL, Lee MA, Finkemeier CG, et al. Operative stabilization of supracondylar femur fractures above total knee arthroplasty: a comparison of four treatment methods. J Arthroplasty 2003;18(7):834. 15. Hou Z, Bowen TR, Irgit K, et al. Locked plating of periprosthetic femur fractures above total knee arthroplasty. J Orthop Trauma 2012;26(7):427. 16. Bong MR, Egol KA, Koval KJ, et al. Comparison of the LISS and a retrograde-inserted supracondylar intramedullary nail for fixation of a periprosthetic distal femur fracture proximal to a total knee arthroplasty. J Arthroplasty 2002;17:876.
Please cite this article as: Meneghini RM, et al, Modern Retrograde Intramedullary Nails Versus Periarticular Locked Plates for Supracondylar Femur Fractures After Total Knee Arthr..., J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.01.025
