Periprosthetic fractures around total knee arthroplasties (TKAs) are well documented in the literature, with rates ranging from 0.3% to 2.5% after primary TKA and 1.6% to 38% following revisions.1–5 Of these, supracondylar femoral fractures are the most common, with periprosthetic tibial fractures occurring much less frequently, with reported rates of 0.4% to 1.7%.1,3,6–9 With more than 300,000 TKAs performed annually in the United States,10 and this number continuing to rise as activity demands increase and implant and patient survival improve, periprosthetic fractures of the tibia will likely rise in incidence.2,3,11–13 Contrasted with other fractures, periprosthetic fractures are often difficult to treat and associated with significant morbidity. They are complicated by such factors as poor bone stock, previous soft tissue injury, implant loosening and instability, and preexisting implant and cement, which may impede successful reduction and placement of fixation devices.1,9,11 These factors pose significant reconstructive challenges to the treating surgeon and pre-dispose to malunion or nonunion.9,14 Due to the rarity of periprosthetic fractures of the tibia, there is a paucity of information concerning clinical and radiographic outcomes following open reduction and internal fixation (ORIF) of these complicated fractures.
The current literature largely bases treatment of periprosthetic tibial fractures on recommendations made by Felix et al7 in 1997. From their review of 102 periprosthetic tibial fractures after TKA at a tertiary care center from 1970 to 1995, the Mayo classification system was created.7 This classification consists of 4 major types (I–IV), with each having 3 associated subtypes (A, B, and C). Type I is a tibial plateau fracture of the tibial tray-to-bone interface. Type II represents a fracture adjacent to the stem. Type III designates a fracture distal to the prosthesis. Type IV identifies a tibial tubercle fracture. The associated subtypes consist of a well-fixed prosthesis (A), a loose prosthesis (B), and intraoperative fracture (C). The vast majority of the fractures in their series were type I (60%), which is much less common currently due to improvements in component design.7,10 Moreover, treatments of these fractures are predominantly focused on stability of the implant and displacement of the fracture, which govern whether nonoperative or operative treatment should be used. With modern implant designs, improved outcomes with operative stabilization in the setting of stable implants with varying degrees of displacement for the less reported type IIA and IIIA fractures are largely unknown because the more recent literature has been limited to reviews, expert opinions, and case reports. The most notable recent advance in the literature regarding the injury is the Unified Classification System (UCS), which is essentially a modification of the original Vancouver classification of periprosthetic hip fractures.6
The current study presents the results of modern internal fixation methods of stable periprosthetic tibial fractures around TKA. In addition, the authors sought to determine whether patients with periprosthetic tibial fractures had significant risk of non-union as well as decreased mobility status following operative treatment of these injuries. Their hypothesis was that these stable Felix type IIA and IIIA, or UCS type B1 and C, fractures would be associated with a significant decrease in ambulation. The authors also sought to determine whether there were any specific identifiable comorbidities or other factors contributing to poorer outcomes among those patients who had periprosthetic tibial fractures.
Materials and Methods
Institutional review board approval was obtained prior to initiation of the study. A retrospective review of a prospectively collected database revealed 19 patients who had a periprosthetic fracture of the tibia around a TKA from 2008 to 2014. All 19 patients underwent operative fixation by 1 of 5 fellowship-trained orthopedic trauma surgeons. Patients were followed at regular intervals with clinical and radiographic evaluation, and the authors not involved with their initial care retrospectively reviewed these data. Four of these patients were lost to follow-up before 1 year, and 1 patient died in the hospital as a result of the constellation of injuries; these 5 were excluded from final analysis, leaving a total of 14 patients for analysis.
Each of the 19 patients underwent fixation using plate and screw instrumentation according to the injury pattern as well as soft tissue consideration. All fixation constructs included a single plate and locking and nonlocking screws in a hybrid construct concept (Figures 1–2); there was no use of cables, allograft, or other fixation adjunct products in this series. All patients were allowed immediate range of motion (ROM) of all joints of the lower extremity postoperatively, with weight bearing proceeding according to surgeon preference but not beginning earlier than 4 weeks postoperatively in this series.
Anteroposterior (A) and lateral (B) radiographs of a 66-year-old woman with a previous total knee arthroplasty who presented after a fall and had sustained a Felix type IIIA fracture. The implant-stable Felix type IIIA fracture can be seen.
Anteroposterior (A) and lateral (B) radiographs of the 66-year-old woman of Figure 1 at 14 months after stabilization with locking screw and plate technology with a hybrid combination of locking and nonlocking screws.
Statistical analysis was performed, with means, ranges, and confidence intervals calculated for continuous variables and compared using the Student's t test. Frequencies were calculated for continuous variables and compared using Fisher's exact test for increased accuracy in small proportion analysis. P<.05 was set as significant, with a trend defined as P being between .05 and .1.
Table 1 presents the demographics and overall basic fracture data. The mean age of the patients was 71.3 years. There were 35.8% male patients, the mean American Society of Anesthesiologists classification was 3.2, and 35.7% were open injuries. All fractures were either type IIA or IIIA by the Felix classification system and B1 or C according to the UCS. A stem was present in 50% of the tibial components.
Patient Demographic and Injury Data
Table 2 presents the surgical and outcome data for the 14 patients within the series, with a mean follow-up of 25.1 months. All fractures were plated using locking screw and plate technology with a hybrid combination of locking and nonlocking screws. Of the 14 patients who were followed for the entire course of treatment, 11 (78.6%) obtained fracture union with the first surgery. Three patients required additional surgery to achieve fracture union, 2 of whom successfully achieved union, for a total union rate of 92.9%. There was only 1 (7.1%) deep infection noted, but 3 (21.4%) of the patients had wound complications that required supplemental tissue coverage. The authors were unable to find any statistically significant differences regarding open fractures and non-union (P=.26), but all of the wound complications regarding skin grafting or flap coverage were originally open injuries (P=.03). The mean postoperative knee ROM was 1.4° to 108.6°, but preoperative ROM was unable to be obtained due to the nature of the injuries.
Patient Surgical and Outcome Data
Periprosthetic tibial fractures around TKAs are a rare but complex injury posing several reconstructive challenges for the operating surgeon. As the use of arthroplasty procedures continues to rise, the incidence of periprosthetic fractures is expected to increase. Understanding implant stability, fracture displacement, and available treatment modalities is imperative to successful outcomes. This is further complicated by treatment guidelines structured around outdated component designs. With newer design technology, the adjuvant use of locking plates in the presence of periprosthetic tibial fractures with a stable implant is unknown. This study has reported the outcomes of operative intervention with internal fixation of implant stable periprosthetic tibial fractures. These were previously recommended to be managed nonoperatively with a cast or brace if the fracture was nondisplaced or minimally displaced, and with closed reduction and casting or ORIF if displaced.7 The authors have reported successful fracture union in 78.6% of patients following initial surgery and in 92.9% of patients after supplemental surgery. Mean time to union was 8.4 months, but postoperative knee ROM was variable.
Prevention of periprosthetic tibial fractures following TKA is ideal. Meticulous surgical technique to minimize bone loss and implant loosening that may subsequently lead to instability, addressing patient risk factors such as osteopenia, and using modern implant designs should reduce this risk.1,2 Moreover, axial malalignment and component malpositioning have been associated with a higher risk of tibial fracture.3 Bone loss and perioperative osteolysis are concerns for postoperative periprosthetic fractures, and some surgeons advocate for routine use of bisphosphonates in all patients undergoing TKA.9,15,16
Despite preventive measures, periprosthetic tibial fractures may still occur. They may occur intraoperatively with skeletal traction, reaming, trialing of components, cementation, or final implant insertion; during revision arthroplasty; or in the postoperative period following trauma or stress.3,8,9,13,17 More commonly, TKA periprosthetic fractures occur following trauma. Several risk factors have been identified, including osteopenia, age older than 70 years, chronic steroid use, inflammatory arthropathies, female sex, stress shielding, revision arthroplasty, and neurological disorders.1–3,10,11,13,18–20
Treatment of these complex fractures is based on previous classifications. The Mayo classification, as popularized by Felix et al,7 largely reported on type I fractures, uncommon currently secondary to improved implant design.10 In their review of 102 periprosthetic tibial fractures, 81% occurred postoperatively. Type I, II, III, and IV fractures occurred in 61, 22, 17, and 2 patients, respectively. Of these, only 4 of the 22 type II fractures were subtype A, all of which were treated with casting or partial weight bearing, and 1 required closed reduction. Fifteen of the 17 type III fractures were subtype A; 2 of these required closed reduction, and 1 that was spiral in nature without evidence of union was treated with ORIF after 3 months. Treatment recommendations were primarily focused on stability of the implant and degree of fracture displacement: nonoperative for stable implants with nondisplaced or minimally displaced fractures (type IIA and IIIA) and operative for unstable implants or irreducible displaced fractures, respectively.2,3,7 Unlike the patients in the current study who had well-fixed implants (subtype A per Felix classification), those with loose implants may fare better with revision arthroplasty.1,3,11 Additionally, diaphysis-engaging tibial intramedullary stems, ORIF of fracture fragments, structural allograft, or tumor prosthesis may be indicated.3,10,21 With the availability of contemporary TKA implants, as well as modern osteosynthesis with locking periarticular plates, union rates and outcomes for patients treated with operative fixation of type IIA and IIIA fractures are unreported in the literature. With heavy emphasis on maximized treatment outcomes, results following this treatment modality are important to the current literature.
The current study had several limitations. First, ROM was variable and depended on many preinjury variables, as well as the fracture and postoperative recovery. Unfortunately, due to poor documentation regarding pre- and postoperative ambulatory functional status (ie, community ambulator, requiring cane, requiring walker, or wheelchair bound), the authors were unable to make any inferences about functional status following operative stabilization of periprosthetic tibial fractures based on the current series. Second, despite publications inferring an increased risk of malunion or nonunion with periprosthetic TKA fractures,9 to the authors' knowledge, no study has reported outcomes or union rates following the treatment recommendations established by Felix et al7 for type IIA and IIIA fractures. Therefore, although the current study provides information regarding union rates in these fractures treated with operative stabilization, there is no literature available to compare the current outcomes with those for nonoperative treatment; thus, the adjuvant use of internal fixation in these fractures remains speculative currently. Additionally, despite being the case series with the largest patient population reported with Felix type IIA and IIIA fractures since the original publication by Felix, the current study had a small patient population. A multicenter retrospective review would represent an improvement in study design and allow for a larger patient population. Future research involving direct comparison of nonoperative and operative management of these fractures is needed. Finally, there were multiple complications noted in this series, which could attest to the complexity of the injuries and support the selection of nonoperative management over operative management in certain instances. Larger prospective, comparative studies would increase the understanding of the prevalence of complications between nonoperative and operative management of these rare fractures.
Periprosthetic tibial fractures are a rare entity in orthopedics, and there remains a paucity of literature available regarding them. In fact, most of the current recommendations are based on the work of Felix et al7 from 20 years ago. Operative intervention may improve results over the current nonoperative recommendations; however, larger study populations are necessary to determine the most appropriate management with current total knee prostheses and osteosynthesis technology. Surgeons must recognize that these injuries are not benign, are typically seen in patients with multiple comorbidities, and can result in lengthy times to union; open injuries are associated with increased rates of wound complications as well.
- Agarwal S, Sharma RK, Jain JK. Periprosthetic fractures after total knee arthroplasty. J Orthop Surg (Hong Kong). 2014;22(1):24–29. doi:10.1177/230949901402200108 [CrossRef] PMID:24781608
- Burnett RS, Bourne RB. Periprosthetic fractures of the tibia and patella in total knee arthroplasty. Instr Course Lect. 2004;53:217–235. PMID: 15116616
- Chimutengwende-Gordon M, Khan W, Johnstone D. Recent advances and developments in knee surgery: principles of periprosthetic knee fracture management. Open Orthop J. 2012;6(1):301–304. doi:10.2174/1874325001206010301 [CrossRef] PMID:22888380
- Healy WL, Siliski JM, Incavo SJ. Operative treatment of distal femoral fractures proximal to total knee replacements. J Bone Joint Surg Am. 1993;75(1):27–34. doi:10.2106/00004623-199301000-00005 [CrossRef] PMID:8419387
- Meek RM, Norwood T, Smith R, Brenkel IJ, Howie CR. The risk of peri-prosthetic fracture after primary and revision total hip and knee replacement. J Bone Joint Surg Br. 2011;93(1):96–101. doi:10.1302/0301-620X.93B1.25087 [CrossRef] PMID:21196551
- Duncan CP, Haddad FS. The Unified Classification System (UCS): improving our understanding of periprosthetic fractures. Bone Joint J. 2014;96-B(6):713–716.
- Felix NA, Stuart MJ, Hanssen AD. Periprosthetic fractures of the tibia associated with total knee arthroplasty. Clin Orthop Relat Res. 1997;345:113–124. doi:10.1097/00003086-199712000-00016 [CrossRef] PMID:9418628
- Rand JA, Coventry MB. Stress fractures after total knee arthroplasty. J Bone Joint Surg Am. 1980;62(2):226–233. doi:10.2106/00004623-198062020-00009 [CrossRef] PMID:7358754
- Yoo JD, Kim NK. Periprosthetic fractures following total knee arthroplasty. Knee Surg Relat Res. 2015;27(1):1–9. doi:10.5792/ksrr.2015.27.1.1 [CrossRef] PMID:25750888
- Parvizi J, Jain N, Schmidt AH. Periprosthetic knee fractures. J Orthop Trauma. 2008;22(9):663–671. doi:10.1097/BOT.0b013e31816ed989 [CrossRef] PMID:18827599
- Della Rocca GJ, Leung KS, Pape HC. Periprosthetic fractures: epidemiology and future projections. J Orthop Trauma. 2011;25 (suppl 2):S66–S70. doi:10.1097/BOT.0b013e31821b8c28 [CrossRef] PMID:21566478
- Kim KI, Egol KA, Hozack WJ, Parvizi J. Periprosthetic fractures after total knee arthroplasties. Clin Orthop Relat Res. 2006;446(446):167–175. doi:10.1097/01.blo.0000214417.29335.19 [CrossRef] PMID:16568003
- Sarmah SS, Patel S, Reading G, El-Husseiny M, Douglas S, Haddad FS. Periprosthetic fractures around total knee arthroplasty. Ann R Coll Surg Engl. 2012;94(5):302–307. doi:10.1308/003588412X13171221592537 [CrossRef] PMID:22943223
- Cordeiro EN, Costa RC, Carazzato JG, Silva JS. Periprosthetic fractures in patients with total knee arthroplasties. Clin Orthop Relat Res. 1990;(252):182–189. doi:10.1097/00003086-199003000-00027 [CrossRef] PMID:2302883
- Wang CJ, Wang JW, Weng LH, Hsu CC, Huang CC, Chen HS. The effect of alendronate on bone mineral density in the distal part of the femur and proximal part of the tibia after total knee arthroplasty. J Bone Joint Surg Am. 2003;85(11):2121–2126. doi:10.2106/00004623-200311000-00009 [CrossRef]
- Soininvaara TA, Jurvelin JS, Miettinen HJ, Suomalainen OT, Alhava EM, Kröger PJ. Effect of alendronate on periprosthetic bone loss after total knee arthroplasty: a one-year, randomized, controlled trial of 19 patients. Calcif Tissue Int. 2002;71(6):472–477. doi:10.1007/s00223-002-1022-9 [CrossRef] PMID:12370800
- Alden KJ, Duncan WH, Trousdale RT, Pagnano MW, Haidukewych GJ. Intraoperative fracture during primary total knee arthroplasty. Clin Orthop Relat Res. 2010;468(1):90–95. doi:10.1007/s11999-009-0876-9 [CrossRef] PMID:19430855
- Platzer P, Schuster R, Aldrian S, et al. Management and outcome of periprosthetic fractures after total knee arthroplasty. J Trauma. 2010;68(6):1464–1470. doi:10.1097/TA.0b013e3181d53f81 [CrossRef] PMID:20539190
- Johnston AT, Tsiridis E, Eyres KS, Toms AD. Periprosthetic fractures in the distal femur following total knee replacement: a review and guide to management. Knee. 2012;19(3):156–162. doi:10.1016/j.knee.2011.06.003 [CrossRef] PMID:21741844
- Dennis DA. Periprosthetic fractures following total knee arthroplasty. Instr Course Lect. 2001;50:379–389. PMID: 11372337
- Hanssen AD, Stuart MJ. Treatment of periprosthetic tibial fractures. Clin Orthop Relat Res. 2000;(380):91–98. PMID:11064978
Patient Demographic and Injury Data
|Age, mean±SD (range), y||71.3±13.2 (52–89)|
|Male sex, No.||5 (35.8%)|
|American Society of Anesthesiologists classification, mean±SD (range)||3.2±0.8 (2–4)|
|Diabetes mellitus, No.||4 (28.6%)|
|Tobacco use, No.||2 (14.3%)|
|Open fracture, No.||5 (35.7%)|
|Unified Classification System classification, No.|
| B1||4 (28.6%)|
| C||10 (71.4%)|
|Felix classification, No.|
| IIA||4 (28.6%)|
| IIIA||10 (71.4%)|
|Stemmed tibial component, No.||7 (50%)|
Patient Surgical and Outcome Data
|Operative time, mean±SD (range), min||111.6±30.6 (71–197)|
|Implant used, No.|
| 3.5-mm anterolateral locking proximal tibial plate||4 (28.6%)|
| 4.5-mm anterolateral locking proximal tibial plate||3 (21.4%)|
| Medial locking distal tibial plate||5 (35.7%)|
| Metaphyseal 3.5-/4.5-mm plate||2 (14.3%)|
|Follow-up, mean±SD (range), mo||25.1±18.4 (4.7–56.0)|
|Fracture union, No.||11 (78.6%)|
|Time to union, mean±SD (range), mo||8.4±3.4 (4–13)|
| Wound issues requiring coverage||3 (21.4%)|
| Nonunion||2 (14.3%)|
| Malunion||1 (7.1%)|
| Deep infection||1 (7.1%)|
| Deep venous thrombosis/pulmonary embolism||0|
|Postoperative knee range of motion (flexion-extension)||1.4°–108.6°|