- Email: [email protected]
Intramedullary Compression Arthrodesis of the Knee
The Journal of Arthroplasty Vol. 20 No. 1 2005
Intramedullary Compression Arthrodesis of the Knee Early Experience With a New Device and Technique David A. McQueen, MD,*y Francis W. Cooke, PhD,* and Dustan L. Hahn, MS*
Abstract: The irretrievably failed total knee arthroplasty is the primary indication for knee arthrodesis. Because this difficult condition is relatively rare, an intramedullary arthrodesis system was developed which requires minimal surgeon experience for successful use. The new system called the Wichita Fusion Nail was implanted by a single surgeon in 13 consecutive patients: 11 for arthrodesis alone, 1 for stabilization of a supracondylar fracture nonunion, and 1 for arthrodesis coupled with a supracondylar fracture nonunion. All arthrodesis attempts were successful. The average fusion time was 15.2 weeks except for 2 infected delayed arthrodeses. Both fracture nonunions persisted and went on to amputation. The WFN provides a simple arthrodesis system with minimal technique dependence and a high potential for success. Key words: knee arthrodesis, knee fusion, intramedullary rod. n 2005 Elsevier Inc. All rights reserved.
knee patient, successful arthrodesis provides a sound sensate limb that permits ambulation and the performance of daily activities at an acceptable level of efficiency [10]. The advantages and disadvantages of various arthrodesis techniques have been reviewed extensively [11-13] starting with the landmark publication of Key [14] in 1932. Variable degrees of success were achieved with onlay grafts and intramedullary (IM) pins until the advent of failed TKA introduced a new set of problems associated with shortened limbs, reduced bone stock, and compromised bony apposition [11]. Whereas current success rates of 80% to 100% are reported for primary arthrodesis in small select patient populations [1,7,8,15,16], clinical results for larger populations with less selective inclusion criteria, that is, those performed secondarily to failed infected TKA, continue to produce fusion rates of 50% [3] to 80% [9,13,17,18]. The objective of the present study was to document the surgical procedure and the level of success achieved with the Wichita Fusion Nail (WFN) in
The irretrievably failed total knee arthroplasty (TKA) is the predominate indication for knee arthrodesis [1-5]. When successful, arthrodesis is very effective in restoring patient function and mobility [1,2,5-7]. In addition, when fusion is achieved, it is remarkable for its ability to relieve pain and for its low reinfection rate [5,6,8,9] in a patient group that has typically endured a long history of previously failed surgeries. In many cases, the alternative to arthrodesis may be an above-knee amputation [4,7]. In the severely compromised
From the *Orthopaedic Research Institute, Inc, Wichita, Kansas, and the y Department of Surgery, Section of Orthopaedics, University of Kansas-Wichita, Wichita, Kansas. Submitted February 26, 2004; accepted July 6, 2004. Benefits or funds were received in partial or total support of the research material described in this article from Stryker Orthopaedics. Reprint requests: Francis W. Cooke, PhD, Orthopaedic Research institute, 929 North St Francis Street, Wichita, KS 67214. n 2005 Elsevier Inc. All rights reserved. 0883-5403/04/2001-0012$30.00/0 doi:10.1016/j.arth.2004.09.022
72
Intramedullary Compression Arthrodesis ! McQueen et al 73
effecting arthrodesis after irretrievable failure of TKA and similar types of severe knee pathology.
Materials and Methods The WFN consists of femoral and tibial IM rods, a split connecting nut, and 4 transverse locking screws (Fig. 1). Development of the device was undertaken with the intent of circumventing the limitations of earlier devices and to provide a simple and dependable system for treating the irretrievably failed TKA. To meet these objectives, 9 design criteria were established from which the final design evolved (Table 1). The overarching objective was to develop a system for the treatment of the irretrievably failed TKA that would provide a very high potential for success in the infrequent circumstance where arthrodesis represents the preferred therapeutic option. This study describes the results of a single surgeon in his first 13 attempts to use the WFN and focuses on the rate of successful arthrodesis as well as the time required to achieve fusion. It was not an objective of this study to document the long-term outcomes of these fusions once clinical arthrodesis had been achieved. The subjects in this study included 13 patients, each of whom was nonambulatory without some form of aid, such as crutches, because of a completely dysfunctional knee. These patients were treated in consecutive order by a single orthopedic surgeon in private practice in a mediumsized city in the American Midwest. Because none of the patients could bear weight on the affected
Fig. 1. Tibial and femoral components of WFN together with compression nut and cross-locking screws.
Table 1. Design Objectives for the WFN 1. 2. 3. 4. 5. 6. 7. 8. 9.
Excellent bending rigidity in all planes Excellent rotational stability Intraoperative compression across the osteotomized surfaces Implantation through a single periarticular incision Ease of insertion without the use of intraoperative x-ray imaging Immediate postoperative mobilization with dynamic compression Good patient acceptance Fatigue resistance adequate to tolerate ambulatory loads for up to 10 million cycles* in cases of delayed union Possibility of device removal without disruption of the union in cases of trauma or infection *Confirmed in preclinical mechanical tests.
leg before the attempted arthrodesis, it was not possible to evaluate them using any of the standard knee function tests, such as the Hospital for Special Surgery Knee Scoring System or the Arthritis Measurements Scale. Postoperative evaluation was based on 5 factors: (1) the absence of pain upon weight bearing, (2) the patients’ acceptance of the inflexible limb, (3) the patients’ ability to walk for at least a short distance without the aid of crutches, (4) radiographic evidence of bony (trabecular) bridging of the osteotomy gap, and (5) the mechanical stability of the fusion upon physical examination. If a positive score was achieved for all 5 criteria, the fusion was rated 5 and was considered an unqualified success. A more rigorous statistical assessment could not be justified because of the small number of cases available and the broad range of diagnostic and demographic variables involved.
Fig. 2. Approximation of the osteotomy surfaces after insertion of the tibial component into the femoral component. Note that the compression nut is in place, and the wrench is positioned to advance the nut into the femoral component.
74 The Journal of Arthroplasty Vol. 20 No. 1 January 2005 locking screws without the need for radiographic image control. Before screw placement is completed, care is taken to ensure proper rotational alignment of the lower limb and foot. After screw placement in each rod, the components are approximated by inserting the connecting shaft of the tibial rod into the femoral component (Fig. 2). The compression nut is placed over the tibial shaft and tightened with a wrench via the anterior window. When the nut is fully tightened, the femoral and tibial osteotomy surfaces are placed in compressive apposition. A unique feature of the WFN design is that the tibial shaft is free to move in a retrograde fashion within the femoral rod even after intraoperative compression has been achieved. This arrangement permits dynamic compression of the fusion during weight bearing without compromising the initial compression. The surgical procedure is completed by replacing the bone blocks and closing the incision. Fig. 3 is a radiograph taken 16 weeks postimplantation showing a well-united arthrodesis. Fusion is evidenced by the presence of bony trabeculation bridging the osteotomy gap. Fig. 3. Radiograph of a fused knee 16 weeks after insertion of a WFN. Evidence of the intra-arthrodesis space has been obliterated by trabecular growth across the fusion interface.
Implantation of the WFN is accomplished through a conventional anterior approach to the knee. The medullary canals are reamed, and transverse osteotomies are prepared at the distal femur and proximal tibia. Anterior bone blocks are also removed from the femur and tibia for access to the joint space during tightening of the compression nut later in the procedure. The tibial and femoral IM rods are inserted, and a bU-armQ drill guide is attached to the rod ends. This drill guide permits percutaneous placement of the transverse
Results Thirteen consecutive WFN implantations were performed by a single surgeon over the course of 55 months. All 13 surgeries (13 patients) were unilateral procedures. The demographic data for this patient group are presented in Table 2. The population was made up of 9 (69%) women with an average age of 62.7 years and 4 (31%) men with an average age of 61.2 years. The average age overall was 62.2 years with an SD of F 27.3 years, which reflected the broad range of patient ages of 24 to 90 years. The range of patient body weights was also wide, extending from 55 kg (120 lb) to 148 kg (325 lb). The previous treatments and diagnoses leading to fusion with the WFN are listed in Table 3. Infected
Table 2. Sex, Age, and Body Weight of Patients Receiving WFN Sex
F M Total F + M
Weight N
%
Nw
Average Weight (kg)
9 4 13
69 31
8 4 12
97.3 92.5 95.7
Age
FSD (kg)
Range (kg)
35.3 – 32.2
55-148 73-136 55-148
Na
Average Age (y)
FSD (y)
Range (y)
9 4 13
62.7 61.2 62.2
19.5 – 27.3
33-90 24-89 24-90
N indicates number of patients; Nw, number of patients for whom weight data were available; Na, number of patients for whom age data were available.
Intramedullary Compression Arthrodesis ! McQueen et al 75 Table 3. Diagnosis and Treatment History of Patients Before Receiving WFN
Sex F M Total F+M Percent of 13 total patients (%)
Infected TKA
Failed TKA
Fracture Nonunion
5 2 7 54
2 1 3 23
1 1 2 15
TKA was the most common diagnosis and accounted for 54% (7/13) of the predisposing conditions. The remaining diagnoses included aseptic failed TKA 23% (3/13), nonunion of periarticular fractures 15% (2/13), and arthritis 8% (1/13). As one would expect, patients receiving the WFN, when taken as a group, had undergone a large number of previous knee surgeries the average being 3.8 (range, 1-7) procedures per patient. No correlation was found between the number of previous surgeries and any of the demographic variables or primary diagnoses. The fracture nonunion patients included 1 who had a successful prior knee arthrodesis but subsequently sustained a supracondylar fracture close to the fusion site. In this case, the WFN was used in an attempt to facilitate healing of the fracture, not to achieve arthrodesis. The remaining fracture nonunion was associated with extensive knee pathology. The WFN was used in this case to stabilize both the fracture and the joint with the intention of facilitating fusion at both locations. Fusion Rates Of the 13 cases included in this study, 11 involved attempts at arthrodesis of the knee in the absence of collateral pathology. The 2 remaining cases both involved attempts to treat supracondylar fracture nonunions. In one case, an attempt was made to fuse the knee and to simultaneously fuse an associated supracondylar fracture nonunion. In the second case, an attempt was made to effect union of a persistent supracondylar fracture nonunion in a patient who had previously experienced a successful knee arthrodesis. Thus, the total patient cohort of 13 included 12 cases of attempted arthrodesis, 1 of which involved a supracondylar fracture. In each arthrodesis case, the assessment of union was based on the 5 criteria listed in Materials and Methods. According to this assessment scale, all 12 arthrodesis attempts
Arthritis
Prior Surgeries (range)
Prior Surgeries (average)
Prior Arthrodesis
1 0 1 8
1-7 2-7 1-7 –
3.7 4.0 3.8 –
0 1 1 –
achieved the maximum score of 5 and were therefore considered to be unqualified successes. These successful arthrodeses did include 2 cases of delayed fusion associated with infections. In one case, the patient progressed to full union after 31 weeks, at which point the WFN was removed without disruption of the arthrodesis and the infection was thereafter resolved without further incident. In the second case, union was achieved after 41 weeks, at which time control of the tuberculosis infection had been achieved medically without removal of the rod. Fusion Time Excluding the delayed unions, the average time to achieve arthrodesis was 15.2 weeks, with an SD of 4.9 weeks and a range of 8 to 24 weeks. The radiographic appearance shown in Fig. 3 is typical of WFN arthrodeses that achieved union in the time frame of 8 to 24 weeks. These healing times compare favorably with previously reported results [2,5,6,10,12,15,16,19]. If the delayed unions of 31 and 41 weeks duration are included in calculating the length of the arthrodesis period, the average fusion time increases from 15.2 to 18.7 weeks. Because of rigid coaxial alignment of the IM rods with the axes of the femur and tibia, all joints were fused at a neutral angle of flexion/ extension with no valgus or varus angulation. Of necessity, this construct imposed some modification on patient gait patterns, including increases in the hip flexion and extension moments. Although no effort has been made to try to assess the magnitude of these increased moments, Puranen et al [15] have concluded b. . .that the resulting position of extension of the knee and the minor shift in the axis of the lower extremity are not reasons to avoid arthrodesis with an intramedullary rod.Q The conditions of failed TKA and other pathology that predispose these knees for arthrodesis all involved some degree of
76 The Journal of Arthroplasty Vol. 20 No. 1 January 2005 limb shortening. Vahvanen [20] has stated that it is better to fuse the knee in the straight position in cases where the classic flexion and valgus angulation would add unduly to leg length discrepancy. In cases where little leg length discrepancy is present, the WFN can accommodate up to 108 of flexion because of the relatively short femoral and tibial IM rods. In general, postfusion walking distances, with or without assistive devices, were limited only by the patient’s endurance. Complications Of the 12 attempts at knee arthrodesis, no case of nonunion was encountered. Two patients experienced delayed unions of 41 and 31 weeks that were associated with deep infections. In the first case, the infection was resolved with appropriate antituberculous therapy, and the rod was left in place. In the second case, successful fusion was followed by removal of the WFN device and effective antibiotic treatment without disruption of the arthrodesis. Device removal was effected by first extracting the transverse locking screws, then transecting the tibial component through its minimum diameter via the preexisting anterior window. The shorter femoral component was then removed through the window, which can be enlarged slightly if necessary. Next, the remainder of the tibial component was moved in a retrograde fashion into the femoral canal. From there, it was also extracted through the anterior window. It is also worth noting that 7 patients in this study had a primary diagnosis of infected TKA (Table 3). Of this group, 2 patients experienced postarthrodesis infections, which represented a reinfection rate of 28% (2/7). This is comparable to the reinfection rates reported [3] for revised infected TKA procedures, which can be as high as 56% [21]. Another complication secondary to a successful arthrodesis was experienced by a 24-year-old man who had initially achieved arthrodesis at 8 weeks. Approximately 1 year later, he sustained a fracture of his tibia in a motor vehicle accident that also bent the tibial component of the WFN at the fracture site. The bent nail was transected at surgery and removed through the fracture site without disruption of the fusion, and the fracture was reduced with compression using a new WFN tibial component. Union was achieved after 17 weeks. This case represented a single example of successful treatment of a periarticular fracture with the WFN. It should be noted, however, that the fracture was acute with no prior history of delayed
union or nonunion. This represented the second case in the series where a WFN was removed without disruption of an established arthrodesis. In 2 cases, attempts were made to use the WFN to treat supracondylar femoral fractures that involved persistent prior nonunions. In the first case, an attempt was made to use the WFN to fix a supracondylar fracture that was located just above a preexisting knee arthrodesis. This attempt was unsuccessful, and the fracture progressed to a persistent nonunion and, eventually, to an aboveknee amputation. The second fracture nonunion patient had sustained a supracondylar fracture that progressed to a persistent nonunion and destruction of the knee. In this case, the WFN was used in an attempt to fuse both the fracture and the knee at the same time. The knee arthrodesis was successful after 16 weeks, but the femoral fracture never united and the patient was eventually treated by an above-knee amputation.
Discussion For more than 70 years, it has been recognized that providing rigid fixation and compression during arthrodesis could reduce failure rates [5,10,17,22]. Compression plating [23,24] provides both rigidity and compression but is more invasive than other procedures and also obviates dynamic compression. External fixators [3-6,10,22] can provide intraoperative compression but are fraught with a number of well-recognized shortcomings. They are difficult and tedious to apply, are subject to pin tract infection, permit little dynamic compression, and are cumbersome for the patient. IM rods offer good bending rigidity, immediate postoperative weight bearing, and are comfortable for the patient [2,3,5,7,12]. However, the IM systems used previously provide poor rotational stability, lack intraoperative compression capability, and sometimes require multiple surgical incisions for implantation. In addition, some IM rod systems make use of long rods that extend through the entire length of the femur [1]. These long rods are extremely difficult to remove in the event of infection or other untoward event and may also complicate future attempts at ipsilateral total hip arthroplasty if such should ever be required. Long IM rods also carry the potential for contamination of otherwise healthy bone when this procedure is performed for failure of TKA secondary to sepsis. The early experience with WFN arthrodesis indicates that this new device and technique have
Intramedullary Compression Arthrodesis ! McQueen et al 77
been successful in avoiding most of the difficulties associated with the earlier devices and methods. This experience confirms that the WFN system meets its primary design objective, which is to provide a routine surgical procedure for the successful treatment of failed TKA, failed prior knee fusion, or severely degenerated arthritic knees, not excluding cases of prior peri-implant infection. The time to fusion (15.2 weeks) and the rate of prompt fusion (83%, 10/12) compare favorably with the best results reported for attempted knee arthrodesis using other means of stabilization. In addition, the 2 cases of delayed arthrodesis were treated successfully and eventually achieved union, so that all 12 attempts at knee arthrodesis were ultimately successful. Because of the limited number of cases with complications, no statistical correlation can be demonstrated between the occurrence of serious arthrodesis complications and any demographic factors. In this series, 2 patients were ultimately treated with above-knee amputation. Both of these unsuccessful cases involved use of the WFN to treat supracondylar fracture nonunions. In a third case, the WFN was used successfully to treat an acute tibial fracture that occurred distal to an established arthrodesis. Although attempts to stabilize periarticular fractures with the WFN failed in 2 of 3 cases, none of these attempts involved a failed knee arthrodesis. In fact, in the one instance where an attempt was made to fuse both the knee and an adjacent fracture, the knee fusion was successful. These experiences should be taken as strong (although limited) evidence that the WFN is not well suited for use in the fixation of periarticular fractures. It is also worth noting that no attempt was made to use the WFN in cases of ipsilateral ankle or hip disease, severe segmental bone loss, or contralateral amputation. In summary, the WFN provides a relatively simple means for effecting surgical fusion after irreversible failure of TKA and other types of severe knee pathology. The WFN relies on cross-locked IM rods to maintain proximal and distal attachment with excellent bending and rotational rigidity. In addition, it allows weight-bearing loads to be carried through the long bones with a dynamic compression capability in much the same way as a compression hip nail.
Conclusion Based on the results of the first 13 cases reported by a single surgeon, it can be concluded that all
attempts at knee arthrodesis with the WFN were successful. Except for the cases of delayed fusion, the average time for achieving arthrodesis was 15.2 weeks. Serious complications were limited to the 2 delayed arthrodeses associated with deep infection and 2 failures to achieve union of supracondylar femur fractures for which the WFN was used as a fixation device. This initial experience confirmed the expectation that the device can be applied to a variety of patients with minimal complications at surgery and a high potential for successful union. The WFN is indicated for failed TKA, failed prior knee fusion, and severe degenerative arthritis of the knee, not excluding cases of prior infection. When necessary, it can be removed without disrupting an already established fusion. Contraindications include use of the WFN to stabilize periarticular fractures. It would also be prudent to exclude it from use in cases of ipsilateral ankle and hip disease, contralateral amputation, and severe segmental bone loss. These results appear to represent a considerable advance over those reported previously for other devices and techniques of knee fusion secondary to severe complications. The high success rate reported here and elsewhere [25] can be attributed to the close adherence of the final device configuration to the exacting design criteria established for the WFN. This is particularly true of the principal objective: that implantation should rely on conventional surgical techniques only, without exceptional prior experience or training.
Acknowledgment The Wichita Fusion Nail (WFN) was conceived, designed, and bench-tested at the Orthopaedic Research Institute of Wichita, Kan. DAM and FWC are codirectors of the Institute. In addition, DAM, FWC, and DLH are designated as the inventors of the WFN on US Patent No. 5 108 398, April 28, 1992. The device is licensed to Stryker Orthopaedics, Mahwah, NJ, which is exclusively responsible for its manufacture and distribution. The authors thank their colleagues, Timothy M. Badwey, MD, Patricia A. Shands, MD, Michelle A. Klauman, MD, and Elizabeth A. Friis, PhD, for their technical contributions to the development of the WFN. The authors also thank Michael J. Long, PhD, of the Wichita State University who reviewed the manuscript. Finally, we are pleased to acknowledge the support of Stryker Orthopaedics.
78 The Journal of Arthroplasty Vol. 20 No. 1 January 2005
References 1. Arroyo JS, Garvin KL, Neff JR. Arthrodesis of the knee with a modular titanium intramedullary nail. J Bone Joint Surg Am 1997;79:26. 2. Ellingsen DE, Rand JA. Intramedullary arthrodesis of the knee after failed total knee arthroplasty. J Bone Joint Surg Am 1994;76:870. 3. Hanssen AD, Trousdale RT, Osmon DR. Patient outcome with reinfection following reimplantation for the infected total knee arthroplasty. Clin Orthop 1995;321:55. 4. Kramhoft M, Bodtker S, Carlsen A. Outcome of infected total knee arthroplasty. J Arthroplasty 1994;9:617. 5. Vlasak R, Gearen PF, Petty W. Knee arthrodesis in the treatment of failed total knee replacement. Clin Orthop 1995;321:138. 6. Benson ER, Resine ST, Lewis CG. Functional outcome of arthrodesis for failed total knee arthroplasty. Orthopedics 1998;21:875. 7. Weiner SD, Scarborough M, Vander-Griend RA. Resection arthrodesis of the knee with an intercalary allograft. J Bone Joint Surg Am 1996;78:185. 8. Wasielewski RC, Barden RM, Rosenberg AG. Results of different surgical procedures on total knee arthroplasty infections. J Arthroplasty 1996;11:931. 9. Wilde AH. Management of infected knee and hip prostheses. Curr Opin Rheumatol 1993;5:317. 10. Woods GW, Lionberger DR, Tullos HS. Failed total knee arthroplasty. Revision and arthrodesis for infection and noninfectious complications. Clin Orthop 1983;173:184. 11. Damron TA, McBeath AA. Arthrodesis following failed total knee arthroplasty: comprehensive review and meta-analysis of recent literature. Orthopedics 1995;18:361. 12. Jorgensen PS, Torholm C. Arthrodesis after infected knee arthroplasty using long arthrodesis nail. A report of five cases. Am J Knee Surg 1995; 8:110.
13. Wilde AH, Stearns KL. Intramedullary fixation for arthrodesis of the knee after infected total knee arthroplasty. Clin Orthop 1989;248:87. 14. Key JA. Positive pressure in arthrodesis for tuberculosis of the knee joint. South Med J 1932;25:909. 15. Puranen J, Kortelainen P, Jalovaara P. Arthrodesis of the knee with intramedullary nail fixation. J Bone Joint Surg Am 1990;72:433. 16. Fahmy NR, Barnes KL, Noble J. A technique for difficult arthrodesis of the knee. J Bone Joint Surg Br 1984;66:367. 17. Goldberg JA, Drummond RP, Bruce WJ, et al. Huckstep nail arthrodesis of the knee: a salvage for infected total knee replacement. Aust N Z J Surg 1989;59:147. 18. Knutson K, Hovelius L, Lindstrand A, et al. Arthrodesis after failed knee arthroplasty. A nationwide multicenter investigation of 91 cases. Clin Orthop 1984;191:202. 19. Figgie III HE, Brody GA, Inglis AE, et al. Knee arthrodesis following total knee arthroplasty in rheumatoid arthritis. Clin Orthop 1987;224:237. 20. Vahvanen V. Arthrodesis in failed knee replacement in eight rheumatoid patients. Ann Chir Gynaecol 1979;68:57. 21. Bengtson S, Knutson K, Lidgren L. Revision of infected total knee arthroplasty. Acta Orthop Scand 1986;57:489. 22. Rand JA, Bryan RS, Chao EY. Failed total knee arthroplasty treated by arthrodesis of the knee using the Ace-Fischer apparatus. J Bone Joint Surg Am 1987;69:39. 23. Nichols SJ, Landon GC, Tullos HS. Arthrodesis with dual plates after failed total knee arthroplasty. J Bone Joint Surg Am 1991;73:1020. 24. Pritchett JW, Mallin BA, Matthews AC. Knee arthrodesis with a tension-band plate. J Bone Joint Surg Am 1988;70:285. 25. Domingo LJ, Caballero MJ, Cuenca J, et al. Knee arthrodesis with the Wichita Fusion Nail. Int Orthop 2004;28:25.
Intramedullary Compression Arthrodesis of the Knee Early Experience With a New Device and Technique David A. McQueen, MD,*y Francis W. Cooke, PhD,* and Dustan L. Hahn, MS*
Abstract: The irretrievably failed total knee arthroplasty is the primary indication for knee arthrodesis. Because this difficult condition is relatively rare, an intramedullary arthrodesis system was developed which requires minimal surgeon experience for successful use. The new system called the Wichita Fusion Nail was implanted by a single surgeon in 13 consecutive patients: 11 for arthrodesis alone, 1 for stabilization of a supracondylar fracture nonunion, and 1 for arthrodesis coupled with a supracondylar fracture nonunion. All arthrodesis attempts were successful. The average fusion time was 15.2 weeks except for 2 infected delayed arthrodeses. Both fracture nonunions persisted and went on to amputation. The WFN provides a simple arthrodesis system with minimal technique dependence and a high potential for success. Key words: knee arthrodesis, knee fusion, intramedullary rod. n 2005 Elsevier Inc. All rights reserved.
knee patient, successful arthrodesis provides a sound sensate limb that permits ambulation and the performance of daily activities at an acceptable level of efficiency [10]. The advantages and disadvantages of various arthrodesis techniques have been reviewed extensively [11-13] starting with the landmark publication of Key [14] in 1932. Variable degrees of success were achieved with onlay grafts and intramedullary (IM) pins until the advent of failed TKA introduced a new set of problems associated with shortened limbs, reduced bone stock, and compromised bony apposition [11]. Whereas current success rates of 80% to 100% are reported for primary arthrodesis in small select patient populations [1,7,8,15,16], clinical results for larger populations with less selective inclusion criteria, that is, those performed secondarily to failed infected TKA, continue to produce fusion rates of 50% [3] to 80% [9,13,17,18]. The objective of the present study was to document the surgical procedure and the level of success achieved with the Wichita Fusion Nail (WFN) in
The irretrievably failed total knee arthroplasty (TKA) is the predominate indication for knee arthrodesis [1-5]. When successful, arthrodesis is very effective in restoring patient function and mobility [1,2,5-7]. In addition, when fusion is achieved, it is remarkable for its ability to relieve pain and for its low reinfection rate [5,6,8,9] in a patient group that has typically endured a long history of previously failed surgeries. In many cases, the alternative to arthrodesis may be an above-knee amputation [4,7]. In the severely compromised
From the *Orthopaedic Research Institute, Inc, Wichita, Kansas, and the y Department of Surgery, Section of Orthopaedics, University of Kansas-Wichita, Wichita, Kansas. Submitted February 26, 2004; accepted July 6, 2004. Benefits or funds were received in partial or total support of the research material described in this article from Stryker Orthopaedics. Reprint requests: Francis W. Cooke, PhD, Orthopaedic Research institute, 929 North St Francis Street, Wichita, KS 67214. n 2005 Elsevier Inc. All rights reserved. 0883-5403/04/2001-0012$30.00/0 doi:10.1016/j.arth.2004.09.022
72
Intramedullary Compression Arthrodesis ! McQueen et al 73
effecting arthrodesis after irretrievable failure of TKA and similar types of severe knee pathology.
Materials and Methods The WFN consists of femoral and tibial IM rods, a split connecting nut, and 4 transverse locking screws (Fig. 1). Development of the device was undertaken with the intent of circumventing the limitations of earlier devices and to provide a simple and dependable system for treating the irretrievably failed TKA. To meet these objectives, 9 design criteria were established from which the final design evolved (Table 1). The overarching objective was to develop a system for the treatment of the irretrievably failed TKA that would provide a very high potential for success in the infrequent circumstance where arthrodesis represents the preferred therapeutic option. This study describes the results of a single surgeon in his first 13 attempts to use the WFN and focuses on the rate of successful arthrodesis as well as the time required to achieve fusion. It was not an objective of this study to document the long-term outcomes of these fusions once clinical arthrodesis had been achieved. The subjects in this study included 13 patients, each of whom was nonambulatory without some form of aid, such as crutches, because of a completely dysfunctional knee. These patients were treated in consecutive order by a single orthopedic surgeon in private practice in a mediumsized city in the American Midwest. Because none of the patients could bear weight on the affected
Fig. 1. Tibial and femoral components of WFN together with compression nut and cross-locking screws.
Table 1. Design Objectives for the WFN 1. 2. 3. 4. 5. 6. 7. 8. 9.
Excellent bending rigidity in all planes Excellent rotational stability Intraoperative compression across the osteotomized surfaces Implantation through a single periarticular incision Ease of insertion without the use of intraoperative x-ray imaging Immediate postoperative mobilization with dynamic compression Good patient acceptance Fatigue resistance adequate to tolerate ambulatory loads for up to 10 million cycles* in cases of delayed union Possibility of device removal without disruption of the union in cases of trauma or infection *Confirmed in preclinical mechanical tests.
leg before the attempted arthrodesis, it was not possible to evaluate them using any of the standard knee function tests, such as the Hospital for Special Surgery Knee Scoring System or the Arthritis Measurements Scale. Postoperative evaluation was based on 5 factors: (1) the absence of pain upon weight bearing, (2) the patients’ acceptance of the inflexible limb, (3) the patients’ ability to walk for at least a short distance without the aid of crutches, (4) radiographic evidence of bony (trabecular) bridging of the osteotomy gap, and (5) the mechanical stability of the fusion upon physical examination. If a positive score was achieved for all 5 criteria, the fusion was rated 5 and was considered an unqualified success. A more rigorous statistical assessment could not be justified because of the small number of cases available and the broad range of diagnostic and demographic variables involved.
Fig. 2. Approximation of the osteotomy surfaces after insertion of the tibial component into the femoral component. Note that the compression nut is in place, and the wrench is positioned to advance the nut into the femoral component.
74 The Journal of Arthroplasty Vol. 20 No. 1 January 2005 locking screws without the need for radiographic image control. Before screw placement is completed, care is taken to ensure proper rotational alignment of the lower limb and foot. After screw placement in each rod, the components are approximated by inserting the connecting shaft of the tibial rod into the femoral component (Fig. 2). The compression nut is placed over the tibial shaft and tightened with a wrench via the anterior window. When the nut is fully tightened, the femoral and tibial osteotomy surfaces are placed in compressive apposition. A unique feature of the WFN design is that the tibial shaft is free to move in a retrograde fashion within the femoral rod even after intraoperative compression has been achieved. This arrangement permits dynamic compression of the fusion during weight bearing without compromising the initial compression. The surgical procedure is completed by replacing the bone blocks and closing the incision. Fig. 3 is a radiograph taken 16 weeks postimplantation showing a well-united arthrodesis. Fusion is evidenced by the presence of bony trabeculation bridging the osteotomy gap. Fig. 3. Radiograph of a fused knee 16 weeks after insertion of a WFN. Evidence of the intra-arthrodesis space has been obliterated by trabecular growth across the fusion interface.
Implantation of the WFN is accomplished through a conventional anterior approach to the knee. The medullary canals are reamed, and transverse osteotomies are prepared at the distal femur and proximal tibia. Anterior bone blocks are also removed from the femur and tibia for access to the joint space during tightening of the compression nut later in the procedure. The tibial and femoral IM rods are inserted, and a bU-armQ drill guide is attached to the rod ends. This drill guide permits percutaneous placement of the transverse
Results Thirteen consecutive WFN implantations were performed by a single surgeon over the course of 55 months. All 13 surgeries (13 patients) were unilateral procedures. The demographic data for this patient group are presented in Table 2. The population was made up of 9 (69%) women with an average age of 62.7 years and 4 (31%) men with an average age of 61.2 years. The average age overall was 62.2 years with an SD of F 27.3 years, which reflected the broad range of patient ages of 24 to 90 years. The range of patient body weights was also wide, extending from 55 kg (120 lb) to 148 kg (325 lb). The previous treatments and diagnoses leading to fusion with the WFN are listed in Table 3. Infected
Table 2. Sex, Age, and Body Weight of Patients Receiving WFN Sex
F M Total F + M
Weight N
%
Nw
Average Weight (kg)
9 4 13
69 31
8 4 12
97.3 92.5 95.7
Age
FSD (kg)
Range (kg)
35.3 – 32.2
55-148 73-136 55-148
Na
Average Age (y)
FSD (y)
Range (y)
9 4 13
62.7 61.2 62.2
19.5 – 27.3
33-90 24-89 24-90
N indicates number of patients; Nw, number of patients for whom weight data were available; Na, number of patients for whom age data were available.
Intramedullary Compression Arthrodesis ! McQueen et al 75 Table 3. Diagnosis and Treatment History of Patients Before Receiving WFN
Sex F M Total F+M Percent of 13 total patients (%)
Infected TKA
Failed TKA
Fracture Nonunion
5 2 7 54
2 1 3 23
1 1 2 15
TKA was the most common diagnosis and accounted for 54% (7/13) of the predisposing conditions. The remaining diagnoses included aseptic failed TKA 23% (3/13), nonunion of periarticular fractures 15% (2/13), and arthritis 8% (1/13). As one would expect, patients receiving the WFN, when taken as a group, had undergone a large number of previous knee surgeries the average being 3.8 (range, 1-7) procedures per patient. No correlation was found between the number of previous surgeries and any of the demographic variables or primary diagnoses. The fracture nonunion patients included 1 who had a successful prior knee arthrodesis but subsequently sustained a supracondylar fracture close to the fusion site. In this case, the WFN was used in an attempt to facilitate healing of the fracture, not to achieve arthrodesis. The remaining fracture nonunion was associated with extensive knee pathology. The WFN was used in this case to stabilize both the fracture and the joint with the intention of facilitating fusion at both locations. Fusion Rates Of the 13 cases included in this study, 11 involved attempts at arthrodesis of the knee in the absence of collateral pathology. The 2 remaining cases both involved attempts to treat supracondylar fracture nonunions. In one case, an attempt was made to fuse the knee and to simultaneously fuse an associated supracondylar fracture nonunion. In the second case, an attempt was made to effect union of a persistent supracondylar fracture nonunion in a patient who had previously experienced a successful knee arthrodesis. Thus, the total patient cohort of 13 included 12 cases of attempted arthrodesis, 1 of which involved a supracondylar fracture. In each arthrodesis case, the assessment of union was based on the 5 criteria listed in Materials and Methods. According to this assessment scale, all 12 arthrodesis attempts
Arthritis
Prior Surgeries (range)
Prior Surgeries (average)
Prior Arthrodesis
1 0 1 8
1-7 2-7 1-7 –
3.7 4.0 3.8 –
0 1 1 –
achieved the maximum score of 5 and were therefore considered to be unqualified successes. These successful arthrodeses did include 2 cases of delayed fusion associated with infections. In one case, the patient progressed to full union after 31 weeks, at which point the WFN was removed without disruption of the arthrodesis and the infection was thereafter resolved without further incident. In the second case, union was achieved after 41 weeks, at which time control of the tuberculosis infection had been achieved medically without removal of the rod. Fusion Time Excluding the delayed unions, the average time to achieve arthrodesis was 15.2 weeks, with an SD of 4.9 weeks and a range of 8 to 24 weeks. The radiographic appearance shown in Fig. 3 is typical of WFN arthrodeses that achieved union in the time frame of 8 to 24 weeks. These healing times compare favorably with previously reported results [2,5,6,10,12,15,16,19]. If the delayed unions of 31 and 41 weeks duration are included in calculating the length of the arthrodesis period, the average fusion time increases from 15.2 to 18.7 weeks. Because of rigid coaxial alignment of the IM rods with the axes of the femur and tibia, all joints were fused at a neutral angle of flexion/ extension with no valgus or varus angulation. Of necessity, this construct imposed some modification on patient gait patterns, including increases in the hip flexion and extension moments. Although no effort has been made to try to assess the magnitude of these increased moments, Puranen et al [15] have concluded b. . .that the resulting position of extension of the knee and the minor shift in the axis of the lower extremity are not reasons to avoid arthrodesis with an intramedullary rod.Q The conditions of failed TKA and other pathology that predispose these knees for arthrodesis all involved some degree of
76 The Journal of Arthroplasty Vol. 20 No. 1 January 2005 limb shortening. Vahvanen [20] has stated that it is better to fuse the knee in the straight position in cases where the classic flexion and valgus angulation would add unduly to leg length discrepancy. In cases where little leg length discrepancy is present, the WFN can accommodate up to 108 of flexion because of the relatively short femoral and tibial IM rods. In general, postfusion walking distances, with or without assistive devices, were limited only by the patient’s endurance. Complications Of the 12 attempts at knee arthrodesis, no case of nonunion was encountered. Two patients experienced delayed unions of 41 and 31 weeks that were associated with deep infections. In the first case, the infection was resolved with appropriate antituberculous therapy, and the rod was left in place. In the second case, successful fusion was followed by removal of the WFN device and effective antibiotic treatment without disruption of the arthrodesis. Device removal was effected by first extracting the transverse locking screws, then transecting the tibial component through its minimum diameter via the preexisting anterior window. The shorter femoral component was then removed through the window, which can be enlarged slightly if necessary. Next, the remainder of the tibial component was moved in a retrograde fashion into the femoral canal. From there, it was also extracted through the anterior window. It is also worth noting that 7 patients in this study had a primary diagnosis of infected TKA (Table 3). Of this group, 2 patients experienced postarthrodesis infections, which represented a reinfection rate of 28% (2/7). This is comparable to the reinfection rates reported [3] for revised infected TKA procedures, which can be as high as 56% [21]. Another complication secondary to a successful arthrodesis was experienced by a 24-year-old man who had initially achieved arthrodesis at 8 weeks. Approximately 1 year later, he sustained a fracture of his tibia in a motor vehicle accident that also bent the tibial component of the WFN at the fracture site. The bent nail was transected at surgery and removed through the fracture site without disruption of the fusion, and the fracture was reduced with compression using a new WFN tibial component. Union was achieved after 17 weeks. This case represented a single example of successful treatment of a periarticular fracture with the WFN. It should be noted, however, that the fracture was acute with no prior history of delayed
union or nonunion. This represented the second case in the series where a WFN was removed without disruption of an established arthrodesis. In 2 cases, attempts were made to use the WFN to treat supracondylar femoral fractures that involved persistent prior nonunions. In the first case, an attempt was made to use the WFN to fix a supracondylar fracture that was located just above a preexisting knee arthrodesis. This attempt was unsuccessful, and the fracture progressed to a persistent nonunion and, eventually, to an aboveknee amputation. The second fracture nonunion patient had sustained a supracondylar fracture that progressed to a persistent nonunion and destruction of the knee. In this case, the WFN was used in an attempt to fuse both the fracture and the knee at the same time. The knee arthrodesis was successful after 16 weeks, but the femoral fracture never united and the patient was eventually treated by an above-knee amputation.
Discussion For more than 70 years, it has been recognized that providing rigid fixation and compression during arthrodesis could reduce failure rates [5,10,17,22]. Compression plating [23,24] provides both rigidity and compression but is more invasive than other procedures and also obviates dynamic compression. External fixators [3-6,10,22] can provide intraoperative compression but are fraught with a number of well-recognized shortcomings. They are difficult and tedious to apply, are subject to pin tract infection, permit little dynamic compression, and are cumbersome for the patient. IM rods offer good bending rigidity, immediate postoperative weight bearing, and are comfortable for the patient [2,3,5,7,12]. However, the IM systems used previously provide poor rotational stability, lack intraoperative compression capability, and sometimes require multiple surgical incisions for implantation. In addition, some IM rod systems make use of long rods that extend through the entire length of the femur [1]. These long rods are extremely difficult to remove in the event of infection or other untoward event and may also complicate future attempts at ipsilateral total hip arthroplasty if such should ever be required. Long IM rods also carry the potential for contamination of otherwise healthy bone when this procedure is performed for failure of TKA secondary to sepsis. The early experience with WFN arthrodesis indicates that this new device and technique have
Intramedullary Compression Arthrodesis ! McQueen et al 77
been successful in avoiding most of the difficulties associated with the earlier devices and methods. This experience confirms that the WFN system meets its primary design objective, which is to provide a routine surgical procedure for the successful treatment of failed TKA, failed prior knee fusion, or severely degenerated arthritic knees, not excluding cases of prior peri-implant infection. The time to fusion (15.2 weeks) and the rate of prompt fusion (83%, 10/12) compare favorably with the best results reported for attempted knee arthrodesis using other means of stabilization. In addition, the 2 cases of delayed arthrodesis were treated successfully and eventually achieved union, so that all 12 attempts at knee arthrodesis were ultimately successful. Because of the limited number of cases with complications, no statistical correlation can be demonstrated between the occurrence of serious arthrodesis complications and any demographic factors. In this series, 2 patients were ultimately treated with above-knee amputation. Both of these unsuccessful cases involved use of the WFN to treat supracondylar fracture nonunions. In a third case, the WFN was used successfully to treat an acute tibial fracture that occurred distal to an established arthrodesis. Although attempts to stabilize periarticular fractures with the WFN failed in 2 of 3 cases, none of these attempts involved a failed knee arthrodesis. In fact, in the one instance where an attempt was made to fuse both the knee and an adjacent fracture, the knee fusion was successful. These experiences should be taken as strong (although limited) evidence that the WFN is not well suited for use in the fixation of periarticular fractures. It is also worth noting that no attempt was made to use the WFN in cases of ipsilateral ankle or hip disease, severe segmental bone loss, or contralateral amputation. In summary, the WFN provides a relatively simple means for effecting surgical fusion after irreversible failure of TKA and other types of severe knee pathology. The WFN relies on cross-locked IM rods to maintain proximal and distal attachment with excellent bending and rotational rigidity. In addition, it allows weight-bearing loads to be carried through the long bones with a dynamic compression capability in much the same way as a compression hip nail.
Conclusion Based on the results of the first 13 cases reported by a single surgeon, it can be concluded that all
attempts at knee arthrodesis with the WFN were successful. Except for the cases of delayed fusion, the average time for achieving arthrodesis was 15.2 weeks. Serious complications were limited to the 2 delayed arthrodeses associated with deep infection and 2 failures to achieve union of supracondylar femur fractures for which the WFN was used as a fixation device. This initial experience confirmed the expectation that the device can be applied to a variety of patients with minimal complications at surgery and a high potential for successful union. The WFN is indicated for failed TKA, failed prior knee fusion, and severe degenerative arthritis of the knee, not excluding cases of prior infection. When necessary, it can be removed without disrupting an already established fusion. Contraindications include use of the WFN to stabilize periarticular fractures. It would also be prudent to exclude it from use in cases of ipsilateral ankle and hip disease, contralateral amputation, and severe segmental bone loss. These results appear to represent a considerable advance over those reported previously for other devices and techniques of knee fusion secondary to severe complications. The high success rate reported here and elsewhere [25] can be attributed to the close adherence of the final device configuration to the exacting design criteria established for the WFN. This is particularly true of the principal objective: that implantation should rely on conventional surgical techniques only, without exceptional prior experience or training.
Acknowledgment The Wichita Fusion Nail (WFN) was conceived, designed, and bench-tested at the Orthopaedic Research Institute of Wichita, Kan. DAM and FWC are codirectors of the Institute. In addition, DAM, FWC, and DLH are designated as the inventors of the WFN on US Patent No. 5 108 398, April 28, 1992. The device is licensed to Stryker Orthopaedics, Mahwah, NJ, which is exclusively responsible for its manufacture and distribution. The authors thank their colleagues, Timothy M. Badwey, MD, Patricia A. Shands, MD, Michelle A. Klauman, MD, and Elizabeth A. Friis, PhD, for their technical contributions to the development of the WFN. The authors also thank Michael J. Long, PhD, of the Wichita State University who reviewed the manuscript. Finally, we are pleased to acknowledge the support of Stryker Orthopaedics.
78 The Journal of Arthroplasty Vol. 20 No. 1 January 2005
References 1. Arroyo JS, Garvin KL, Neff JR. Arthrodesis of the knee with a modular titanium intramedullary nail. J Bone Joint Surg Am 1997;79:26. 2. Ellingsen DE, Rand JA. Intramedullary arthrodesis of the knee after failed total knee arthroplasty. J Bone Joint Surg Am 1994;76:870. 3. Hanssen AD, Trousdale RT, Osmon DR. Patient outcome with reinfection following reimplantation for the infected total knee arthroplasty. Clin Orthop 1995;321:55. 4. Kramhoft M, Bodtker S, Carlsen A. Outcome of infected total knee arthroplasty. J Arthroplasty 1994;9:617. 5. Vlasak R, Gearen PF, Petty W. Knee arthrodesis in the treatment of failed total knee replacement. Clin Orthop 1995;321:138. 6. Benson ER, Resine ST, Lewis CG. Functional outcome of arthrodesis for failed total knee arthroplasty. Orthopedics 1998;21:875. 7. Weiner SD, Scarborough M, Vander-Griend RA. Resection arthrodesis of the knee with an intercalary allograft. J Bone Joint Surg Am 1996;78:185. 8. Wasielewski RC, Barden RM, Rosenberg AG. Results of different surgical procedures on total knee arthroplasty infections. J Arthroplasty 1996;11:931. 9. Wilde AH. Management of infected knee and hip prostheses. Curr Opin Rheumatol 1993;5:317. 10. Woods GW, Lionberger DR, Tullos HS. Failed total knee arthroplasty. Revision and arthrodesis for infection and noninfectious complications. Clin Orthop 1983;173:184. 11. Damron TA, McBeath AA. Arthrodesis following failed total knee arthroplasty: comprehensive review and meta-analysis of recent literature. Orthopedics 1995;18:361. 12. Jorgensen PS, Torholm C. Arthrodesis after infected knee arthroplasty using long arthrodesis nail. A report of five cases. Am J Knee Surg 1995; 8:110.
13. Wilde AH, Stearns KL. Intramedullary fixation for arthrodesis of the knee after infected total knee arthroplasty. Clin Orthop 1989;248:87. 14. Key JA. Positive pressure in arthrodesis for tuberculosis of the knee joint. South Med J 1932;25:909. 15. Puranen J, Kortelainen P, Jalovaara P. Arthrodesis of the knee with intramedullary nail fixation. J Bone Joint Surg Am 1990;72:433. 16. Fahmy NR, Barnes KL, Noble J. A technique for difficult arthrodesis of the knee. J Bone Joint Surg Br 1984;66:367. 17. Goldberg JA, Drummond RP, Bruce WJ, et al. Huckstep nail arthrodesis of the knee: a salvage for infected total knee replacement. Aust N Z J Surg 1989;59:147. 18. Knutson K, Hovelius L, Lindstrand A, et al. Arthrodesis after failed knee arthroplasty. A nationwide multicenter investigation of 91 cases. Clin Orthop 1984;191:202. 19. Figgie III HE, Brody GA, Inglis AE, et al. Knee arthrodesis following total knee arthroplasty in rheumatoid arthritis. Clin Orthop 1987;224:237. 20. Vahvanen V. Arthrodesis in failed knee replacement in eight rheumatoid patients. Ann Chir Gynaecol 1979;68:57. 21. Bengtson S, Knutson K, Lidgren L. Revision of infected total knee arthroplasty. Acta Orthop Scand 1986;57:489. 22. Rand JA, Bryan RS, Chao EY. Failed total knee arthroplasty treated by arthrodesis of the knee using the Ace-Fischer apparatus. J Bone Joint Surg Am 1987;69:39. 23. Nichols SJ, Landon GC, Tullos HS. Arthrodesis with dual plates after failed total knee arthroplasty. J Bone Joint Surg Am 1991;73:1020. 24. Pritchett JW, Mallin BA, Matthews AC. Knee arthrodesis with a tension-band plate. J Bone Joint Surg Am 1988;70:285. 25. Domingo LJ, Caballero MJ, Cuenca J, et al. Knee arthrodesis with the Wichita Fusion Nail. Int Orthop 2004;28:25.