Management of Periprosthetic Fracture in Unicompartmental Knee Arthroplasty Patients: A Case Series Yew Lok Woo, MBBS, Pak Lin Chin, MBBS, FRCS, Ngai Nung Lo, MBBS, FRCS, Shi-Lu Chia, PhD, Darren Keng Jin Tay, MBBS, FRCS, Seng Jin Yeo, MBBS, FRCS Department of Orthopaedic Surgery, Singapore General Hospital, Singapore Abstract Unicompartmental knee arthroplasty (UKA) has been one of the treatment modality specifically for unicompartmental osteoarthritis of the knee. The advantages of UKA are faster recovery period, shorter length of hospital stay, reduced morbidity, and good functional outcome. However, one of the complications in UKA is periprosthetic fracture where most papers suggested a surgical intervention such as revision to total knee replacement. In our six of experience with UKA surgery from 2005 to 2010, we encountered six periprosthetic fractures out of 966 knees that were operated on among 901 patients. Five patients were treated conservatively by casting and were advised not to bear weight on the affected side. One patient was revised to total knee replacement immediately. Each patient underwent a series of scoring system such as Knee Society Clinical Rating and Oxford Questionnaire to evaluate the outcome of treatment. The results were variable with only one patient showing reasonable improvement by conservative management alone. Another three patients did not show significant functional improvements despite the fractures healing. One patient developed non-union and had to undergo internal fixation to correct the deformity. The patient who underwent a revision to total knee replacement showed significant improvement. We conclude that surgical intervention produced better outcome. Revision to total knee replacement was the preferred treatment. Keywords: Management, Periprosthetic fracture, Unicompartmental knee arthroplasty INTRODUCTION Unicompartmental knee arthroplasty (UKA) is one of the treatment modality for single compartment arthritis of the knee. The proposed advantages of UKA are faster recovery period, shorter length of hospital stay, reduced morbidity, less infection, less thromboembolic event, and good functional outcome when compared to other treatments for single compartment arthritis 1 4. The Oxford Group had proven durability up to 20 and provided good long-term pain relief in elderly patients 5,6. Despite the above advantages, there are still complications that can impact the outcome. These include infection, ligamentous instability, periprosthetic fracture, displacement, and loosening of prosthesis 7,8. Periprosthetic fracture in UKA patients is an uncommon complication. Pandit et al. (2000) reviewed The Swedish Knee Arthroplasty Registry and found that there were less than 1% of periprosthetic fractures in UKA in more than 1000 cases performed and comprised of only 1% cause of revision 9. Most periprosthetic fractures in UKA patients were reported as case reports and there has so far been no consensus about the appropriate treatment for this complication. Intuitively, conversion of fractured UKA to a total knee replacement will be the logical bail out 10. In this paper, we would like to share our experiences on the different treatment options for these fractures to aid surgeons in choosing the best management plan if such events occur. MATERIALS AND METHODS X-rays and medical records of all patients who underwent unicompartmental knee arthroplasty from 1 January 2006 to 31 December 2010 in a single institution registry (Singapore General Hospital) were audited. Patient s demographic data such as gender, age, body mass index (BMI), pre-morbids, knee society clinical rating score 267
(KSCR), Oxford Questionnaires, and the alignment of the knees before surgery, six and two post-surgery were analysed. Patients who sustained a periprosthetic fracture and the mode of treatment were identified. The outcomes of each patient were compared with the general outcome of the population. RESULTS From our registry data, a total of 966 knees were operated among 901 patients over six. Majority were females (721) with female to male ratio of 2.96:1. The average BMI of our patients was 25.4 kg/m 2. Fifty-four patients underwent bilateral UKA at a single operation whereas 11 of them had staged surgeries. All patients were operated by five arthroplasty-trained surgeons. The demographic data is further explained in Table 1. During the six period, a total of six cases of periprosthetic fractures (incidence of 0.0062) were encountered. All of them were females with ages ranging from 58- to 77--old (mean 67.5) and BMI ranging from 19.3 to 40.1 (mean 28.1). of them were known to have osteopaenia prior to operation and half of them were obese. Preoperatively, they had genu varus of 2 6 degrees. Five of them presented with persistent knee pain on ambulation one month after surgery and another one had knee pain after a fall five post- Table 1. Demographic data of unicompartmental knee arthroplasty (UKA) patients in the Singapore General Hospital from 1 January 2006 to 31 December 2010. Number of patients 901 Bilateral UKA 54 Number of knees 966 Staged UKA 11 Female patients 721 Previous HTO* 2 2.96:1 Male patients 243 Number of revisions 13 Average weight 62.5 kg Number of patients with fractures 6 Average body mass index 25.0 kg/m 2 Average duration of hospitalisation 4.3 days *HTO: high tibial osteotomy Fig. 1. X-ray of Patient A with minimally displaced periprosthetic fracture. Fig. 2. Radiograph of Patient B who sustained a periprosthetic fracture a month after surgery. Fig. 3. X-ray of Patient C with mildly displaced periprosthetic fracture. 268
Management of Periprosthetic Fracture in UKA Patients operation. All of them sustained fractures at the medial tibia plateau with mild displacement noted on initial radiographs (Fig. 1, 2, and 3). However, no joint instability or ligamentous laxity was noted in all patients. The five patients who presented a month after surgery were offered either revision to total knee replacement or surgical fixation but they opted not to have surgical intervention initially and were casted for six weeks. Three out of five patients showed no further displacement of the fracture (Fig. 4). Patient B had further displacement (Fig. 5) but the patient still declined any surgery. Patient C developed non-union after being treated conservatively. She underwent medial buttress plate with bone grafting at six after initial surgery (Fig. 6). Patient F who sustained a periprosthetic fracture after a fall at five (Fig. 7) was revised to a total knee replacement with stem and wedge. Fig. 4. Patient A s knee x-ray, three after sustaining fracture. Fig. 5. Repeated x-ray of Patient B, three later. Fig. 6. X-ray of Patient C after undergoing surgery. Fig. 7. Patient F sustained fracture after a fall. Fig. 8. X-ray of Patient F after revision to total knee replacement. 269
Table 2. Basic information of the six periprosthetic fracture patients and comparison of knee society clinical rating between the six patients and the general population. Patients Gender Age BMI Duration of fracture noted post-operative Management Knee Society clinical rating Function score Knee score Mean A Female 62 19.3 1 month Casr 70 80 80 50 81 70 B Female 58 29.5 1 month Cast 55 45 80 70 74 66 C Female 76 24.3 1 month Cast then fixed at 6 post-operative due to non-union 60 5* 75 30 83* 80 70 (60.6) D Female 67 33 1 month Cast 50 35 60 28 54 35 80 (76.9) 80 (61.9) 28 (43.1) 90 (86.2) E Female 77 22.5 1 month Cast 70 70 80 43 72 75 F Female 65 40.1 5 Revision to total knee replacement 35 70 # 80 45 66 # 84 * The six score was done prior to fixation. # The six score was done after revision to total knee replacement. Table 3. Range of motion of the unicompartmental knee arthroplasty patients and their Oxford Knee Questionnaire as compared to the general population. Lower scores in Oxford Knee Questionnaire denote better functional outcome and less pain. Patients Range of motion ( o ) Oxford Knee Questionnaire Mean A 5 to 143 0 to 132-3 to 140 23 21 18 B 1 to 125 0 to 120 4 to 105 22 29 26 C 0 to 129 5* to 139 0 to 145 37 26* 19 7.5 120 (4.9 to 127.5) 7.5 122.5 (4.2 to 125.5) 6 128.5 (2.2 to 129) D 5 to 140 2 to 124 0 to 133 45 33.5 (32.8) 41 19 (19.3) 28 E 15 to 115 14 to 112 7 to 110 28 16 12 17.5 (17.8) F 9 to 110 4 # to 109 4 to 120 46 19 # 20 * The six score was done prior to fixation. # The six score was done after revision to total knee replacement. 81 (67.6) 270
Management of Periprosthetic Fracture in UKA Patients Table 4. The sagittal alignment of the knee for all six patients. Patients erative A 5 O Varus 2 O Valgus 4 O Varus B 2 O Varus 2 O Varus 5 O Valgus C 1 O Varus 5 O Valgus* 10 O Valgus D 2 O Varus 3 O Valgus 5 O Varus E 5 O Varus 4 O Varus 1 O Varus F 6 O Varus 5 O Valgus # 5 O Valgus * The six score was done prior to fixation. # The six score was done after revision to total knee replacement. A lag screw was used to fix the fracture (Fig. 8). Post-operative recovery was uneventful with good functional outcome The functional outcome of all six patients is shown in Tables 2, 3, and 4 (please see overleaf). The knee society clinical rating, Oxford Questionnaire, and the alignment of the knees of all six patients were compared to the mean and median scores of 713 patients out of 901 patients who had completed two follow-up. DISCUSSION Periprosthetic fracture is a rare and potentially disastrous complication for UKA patients. The risk factors include osteoporosis, rheumatoid arthritis, osteomalacia, being female, and obesity 11,12. Majority of the periprosthetic fractures in UKA occurred at the tibial plateau 7,9,13,14. It was postulated that stress fractures may occur during the sagittal cut, excessive removal of bone below the subchondral bone, inadequate preparation of keel slot, and usage of more than three pins to anchor jigs 15 17. These points of weakness can eventually evolve into a complete fracture when the patients start to bear weight after surgery. Periprosthetic fractures can be treated conservatively or surgically depending on the stability of fracture configuration. In the context of UKA patients, there is very little in the literature on the best treatment modality. Common consensus suggests that periprosthetic fracture should be treated surgically, either with a formal revision to total knee replacement or internal fixation of the fracture 18. In this study, we encountered five periprosthetic fractures that were noticed a month after surgery. We postulated that these were likely due to a stress fracture that may have occurred intra-operatively and were unnoticed during the initial post-surgery x-rays. These patients were treated conservatively based on their request. They were all casted for at least six weeks without bearing weight on the affected knee. Four patients fracture united and one developed non-union. The fracture had caused implant malalignment which resulted in varus and valgus deformity as reflected in Table 4. As shown in Tables 2, 3, and 4, only Patient A had a reasonably acceptable outcome in all aspects (improvements in both knee and function scores in KSCR, Oxford Questionnaire, range of motion and alignment) after two of follow-up. The remaining three patients had worsened results in certain entities, such as reduced range of motion or worsened knee score after two. Although casting may save the patient from surgery, prolonged immobilisation may lead to knee stiffness and muscle atrophy. Patient C was treated with plate fixation after developing a non-union. A better outcome was noted with such treatment. However, the alignment of the knee deteriorated over the. This was because plate fixation was only able to prevent further collapsing of the fractured fragment and it was not able to correct the ligamentous insufficiency that may have occurred due to the fracture. Hence, the longevity of such treatment is questionable as it did not prevent disease progression; revision to total knee replacement in this case was unavoidable. On the other hand, Patient F who had a revision to total knee replacement showed better outcomes post-operatively. Decisive intervention with total knee replacement allowed the surgeon to address malalignment, ligamentous laxity, and bone loss during the surgery. This allowed early mobilisation which resulted in improved range of motion of the joint and reduced incidence of disuse muscle atrophy. 271
Although patients with conservative management did show improvement in certain entities in KSCR and Oxford Questionnaire, the results were not convincing and the outcome of Patient A was not reproducible in the four other patients. Nevertheless, mal- or non-union can develop in patients who were treated conservatively as shown in Patient D. Therefore, future UKA patients who develop a fracture post-operatively, revision to total knee replacement will be a better choice as it provides a more predictable positive outcome. In this study, we studied a small cohort of cases and hence no qualitative comparison was carried out. CONCLUSION Periprosthetic fractures in UKA patients are uncommon (incidence of 0.0062 in this study) and conservative management for periprosthetic fracture in UKA patients does not show promising results. Hence, surgical interventions are recommended in such a condition and revision to total knee replacement is more preferred. knee arthroplasty. Acta Orthop Belg 2006;72(3):369 74. 11. Poss R, Ewald FC, Thomas WH, Sledge CB. Complications of total hip-replacement arthroplasty in patients with rheumatoid arthritis. J Bone Joint Surg Am 1976;58(8):1130 3. 12. Lewold S. The Swedish knee arthroplasty study with special reference to unicompartmental prostheses. Lund, Sweden:Lund University;1997. 13. Rudol G, Jackson MP, James SE. Medial tibial plateau fracture complicating unicompartmental knee arthroplasty. J Arthroplasty 2007;22(1):148 50. 14. Kim KT, Lee S, Cho KH, Kim KS. Fracture of the medial femoral condyle after unicompartmental knee arthroplasty. J Arthroplasty 2009;24(7):1143.e21 4. 15. Clarius M, Haas D, Aldinger PR, Jaeger S, Jakubowitz E, Seeger JB. Periprosthetic tibial fractures in unicompartmental knee arthroplasty as a function of extended sagittal saw cuts: An experimental study. Knee 2010;17(1):57 60. 16. Sloper PJ, Hing CB, Donell ST, Glasgow MM. Intra-operative tibial plateau fracture during unicompartmental knee replacement: A case report. Knee 2003;10(4):367 9. 17. Brumby SA, Carrington R, Zayontz S, Reish T, Scott RD. Tibial plateau stress fracture: A complication of unicompartmental knee arthroplasty using 4 guide pin holes. J Arthroplasty 2003;18(6):809 12. 18. Berend KR, George J, Lombardi AV Jr. Unicompartmental knee arthroplasty to total knee arthroplasty conversion: Assuring a primary outcome. Orthopedics 2009;32(9) doi: 10.3928/01477447-20090728-32. REFERENCES 1. Beard DJ, Murray DW, Rees JL, Price AJ, Dodd CA. Accelerated recovery for unicompartmental knee replacement A feasibility study. Knee 2002;9(3):221 224. 2. Newman JH, Ackroyd CE, Shah NA. Unicompartmental or total knee replacement? Five-year results of a prospective, randomized trial of 102 osteoarthritic knees with unicompartmental arthritis. J Bone Joint Surg Br 1998;80(5):862 5. 3. Bengtson S, Knutson K. The infected knee arthroplasty. A 6-year follow-up of 357 cases. Acta Orthop Scand 1991;62(4):301 11. 4. Weale AE, Halabi OA, Jones PW, White SH. Perceptions of outcomes after unicompartmental and total knee replacements. Clin Orthop Relat Res 2001;382:143 53. 5. Price AJ, Svard U. A second decade life table survival analysis of the Oxford Unicompartmental Knee Arthroplasty. Clin Orthop Relat Res 2011;469(1):174 9. 6. Ansari S, Newman JH, Ackroyd CE. St. Georg sledge for medial compartment knee replacement. 461 arthroplasties followed for 4 (1-17). Acta Orthop Scand 1997;68(5)430 4. 7. Yang KY, Yeo SJ, Lo NN. Stress fracture of the medial tibial plateau after minimally invasive unicompartmental knee arthroplasty: A report of 2 Cases. J Arthroplasty 2003;18(6)801 3. 8. Furnes O, Espehaug B, Lie SA, Vollset SE, Engsaeter LB, Havelin LI. Failure mechanisms after unicompartmental and tricompartmental primary knee replacement with cement. J Bone Joint Surg Am 2007;89(3):519 25. 9. Pandit H, Murray DW, Dodd CA, Deo S, Waite J, Goodfellow J. Medial tibial plateau fracture and the Oxford unicompartmental knee. Orthopedics 2007;30 (5suppl):28 31. 10. Van Loon P, de Munnynck B, Bellenmans J. Periprosthetic fracture of the tibial plateau after unicompartmental 272