Where Are We Now?
Treatment of medial compartment arthritis in the knee with unicompartmental knee arthroplasty (UKA) either provides early restoration of mobility, relief of pain, and excellent 10-year survivorship [1, 2, 6] or shows worse survivorship than TKA [4, 5, 7]. An analysis of the reasons for UKA revision in a countrywide survey performed in France found that loosening was the most common cause, followed by progressive arthritis, polyethylene wear, and knee pain . To understand tibial pain as a reason for revision of UKA, numerical and mechanical models have shown increased bone strain in the anterior aspect of the resurfaced tibia when load is applied, implying that this increased bone strain is the cause of pain [9-11]. The current study by Peersman et al.  confirms there is an increase in proximal tibial bone strain after UKA compared with the normal knee, at least in cadaver specimens with a loading protocol and specific mobile- and fixed-bearing UKA implants. The current study describes tibial strain measures recorded during simulated deep squatting. This activity produces a load component in the plane of the tibia together with a small vertical load component. Earlier studies [9-11] used vertical loads during measurement or prediction of tibial strain and found similar increases in proximal tibial strain after UKA was performed. With squatting loads, strain measures in the specific anterolateral aspect of the medial tibia showed the mobile-bearing implant was associated with near-normal strain while the fixed-bearing implant had increased strain in this region. This is probably because the mobile-bearing knee implant does not transfer in-plane loads to the underlying bone.
Where Do We Need To Go?
Studies give few clues as to the cause of tibial pain after UKA. Is tibial pain more common with some UKA designs than with others? Is pain related to a particular patient activity, such as walking, stair climbing or descending, rising from a chair, or even deep squatting? Is it continuous or intermittent? Do “long-term survivors” of UKA experience tibial pain? Do certain activities cause tibial pain more often than others do? Can knee loading measured during activities that produce pain be simulated and incorporated into models so that the stress state of the painful knee is understood more explicitly?
This study by Peersman et al.  continues suggestions that an increase in tibial strain after UKA may be the cause of tibial pain. Clearly, these findings alone do not show that pain and increased strain are related to one another. However, assuming the model results are correct, a permanent change in bone strain in a patient with UKA should cause a change in the density of tibial cancellous bone with time, as described by Wolff’s Law. Detailed radiographic analyses will help to show whether the change in strain state in the tibia causes tibial remodeling. To my knowledge, there is little or no radiographic evidence to show the radiographic density change patterns caused by an increase in tibial strain after UKA. Without a general remodeling profile, we cannot determine whether patients undergoing UKA who have pain experience different bone density changes than patients who do not have pain. Thus, without confirmation by radiographic data, the strain predictions may be in error or the strain increase may not be enough to cause bone remodeling. Patients with painful hip and knee implants often have radiographic changes such as trabecular streaming, spot welds, osteopenia, and interface loosening that explain the pain. With the lack of compelling radiographic data after UKA, perhaps a careful evaluation of bone interfaces when an implant is revised for pain will help us understand whether the implant is simply loose or if strain-related adverse changes in the proximal tibia have occurred.
How Do We Get There?
Different authors using models incorporating different UKA implants subjected to different loading regimens predicted a general increase in tibial strain under load after UKA compared with the native knee [8-10]. The authors further suggest that increased tibial strain is a cause of tibial pain that can lead to revision of the procedure. Confirmation or otherwise of these predictions will require a two-step evaluation.
First, an increase in the state of strain in the proximal tibia should produce a concomitant increase in radiographic density in the same tibial location. To determine whether these increased strain predictions are correct, an analysis of the correlation between the location of the predicted increased strain state and the location of increases in the density of tibial cancellous bone that occur with time after UKA is required. A prospective, quantitative study of density changes could be designed and conducted. However, a simpler retrospective analysis of preoperative and follow-up radiographic images of the proximal tibia in survivors of UKA will show whether density changes take place and do or do not occur in the general locations and patterns predicted by in vitro models.
Second, to determine a correlation between bone remodeling and pain after UKA, carefully designed prospective clinical studies will be required. Unfortunately, published clinical results of the success or failure of UKA are so varied and the results of studies are so different that selected data support preconceived notions about the utility of UKA [1-7]. Published clinical studies tend to report generic measures of pain (such as knee pain or no pain), but generic findings are not helpful in determining whether pain is related to a specific scientific question (for example, does increased tibial strain cause pain?). Perhaps existing studies can help to determine whether there are common factors in the demographics of patients who undergo revision for tibial pain. To determine the relationship between strain and pain, we need more-specific clinical data on patients who report pain (that is, when, how often, where, how much, and what activity). This can be achieved with prospective studies in which the implantation technique, implant type, and patient-reported outcomes and pain questions are carefully designed and controlled. Along with patient outcome data, radiographic data need to be collected at each patient visit so that specific reports of pain at a follow-up visit can be compared with radiographic changes since implantation. If the patient undergoes revision for pain, the radiographic data can show whether the areas of increased strain predicted by the models have densified and the patient can report the pain location. Unfortunately, such a study will require a large patient cohort and will probably require multicenter involvement to achieve a high probability of proving the strain-pain relationship. If a strain-pain relationship is found and the activity causing pain is identified, models can be improved by simulating specific loading associated with the activity that causes the pain.
The finding that increased bone strain is predicted for the tibia after UKA holds for all recipients of UKA, whether there is pain or not. Increased strain may be the cause of proximal tibial pain but is not yet a proven causative factor. Interestingly, some patients who undergo UKA are happy with their knee for many years postoperatively [1, 2, 6]. Thus, the premise that increased tibial bone strain causes pain in some patients with UKA and not others remains speculative. All orthopaedic implants cause local changes in bone strain together with associated bone remodeling. Benign remodeling usually does not cause pain. We need to determine whether the changes in bone strain predicted for UKA produce related changes in local bone density. A careful analysis using bone imaging can show whether patients with tibial pain show signs of adverse tibial remodeling compared with patients who do not have pain. In UKA, the relationship among bone strain, adaptive remodeling, and pain is unknown. Clarification will help to improve the design of tibial implants and the tools used to implant them.
1. Berger RA, Meneghini RM, Jacobs JJ, Sheinhop MB, Della Valle CJ Rosenberg AC, Galante JO. Results of unicondylar arthroplasty at minimum ten years follow-up. J Bone Joint Surg Am. 2005;87:999-1006.
2. Deshmukh RV, Scott RD. Unicompartmental knee arthroplasty: long term results. Clin Orthop Relat Res.
3. Epinette JA, Brunschweiler B, Mertl M, Mole D, Cazenave A, French Society for Hip and Knee. Unicompartmental modes of failure: wear is not the main reason for failure: a multicentre study of 418 failed knee implants. Orthop Traumatol Surg Res. 2012;98:S124-S130.
4. Liddle AD, Judge AJ, Pandit H, Murray DW. Adverse outcomes after total and uncompartmental knee replacement in 101 330 matched patients: a study of data from the National Joint Registry for England and Wales. Lancet. 2014;384:1401-1405.
5. Murray DW, Liddle AD, Dodd CAF, Pandit H Unicompartmental knee arthroplasty: is the glass half full or half empty? Bone Joint J. 2015;97:3-8.
6. Newman J, Pydisetty RV, Ackroyd C. Unicondylar knee or total knee arthroplasty. The fifteen year results of a prospective randomized controlled trial. J Bone Joint Surg Br. 2009;91:983-989.
7. Niinimaki T, Eskelinen A, Makela K, Ohtonen P, Puhto AP, Remes V.Unicompartmental knee arthroplasty survivorship is lower than TKA survivorship: a 27-year Finnish registry study. Clin Orthop Relat Res. 2014;472:1496-1501.
8. Peersman G, Taylan O, Verhaegen J, Anthoniessen L, Slane J, Vanthienen B, van Lenthe H, Heyse T, Scheys L. Does unicondylar knee arthroplasty affect tibial bone strain? A paired cadaveric comparison of fixed- and mobile-bearing designs. Clin Orthop Relat Res. 2020. [Published online ahead of print]. DOI: 10.1097/CORR.0000000000001169
9. Scott CEH, Eaton MJ, Nutton RW, Wade FA, Evans SL, Pankaj P. Metal-backed versus all-polyethylene unicompartmental knee arthroplasty. Proximal tibial strain in an experimentally validated finite element model. Bone Joint Res. 2017;6:22-30.
10. Scott CEH, Eaton MJ, Nutton RW, Wade FA, Pankaj P, Evans SL. Proximal tibial strain in medial unicompartmental knee replacements. A biomechanical study of implant design. Bone Joint J. 2013;95:1339-1347.
11. Simpson DJ, Price AJ, Gulati DW, Gill HS. Elevated proximal tibial strains during unicompartmental knee replacement – a possible cause of pain. Med Eng Phys. 2009;31:752-757.