More than one in four patients who undergo unicompartmental knee arthroplasty, a partial knee replacement procedure, are left with fragments of bone cement around or inside the joint, according to a retrospective study of 321 patients treated at a single centre in China. The research, published in the journal Arthroplasty, found residual cement in 84 of 321 patients, a prevalence of 26.16 percent, a figure the authors describe as likely an underestimate because computed tomography was performed only in patients with symptoms.
The study, led by Zhenggui Yu and colleagues at the First Affiliated Hospital of Fujian Medical University, divided residual cement into two categories. Residual cement extrusion, or RCE, was defined as a radiopaque shadow of cement extending beyond the implant component or the bone cortex on postoperative radiographs. Free cement fragmentation, or FCF, referred to cement that had detached from its original location and was present within the joint cavity. Of the 84 affected patients, 66, or 20.56 percent, had RCE and 18, or 5.61 percent, had FCF.
The clinical picture was strikingly benign for most patients. Every one of the 66 patients with residual cement extrusion was either entirely asymptomatic or experienced only mild discomfort, defined in the study as pain below a visual analogue score of 4 and a knee range of motion of at least 90 degrees. These patients generally did well with observation and conservative treatment such as physical therapy and oral analgesics when needed. The authors argue that because extruded cement so rarely causes trouble, it should be regarded as a clinical phenomenon rather than a clinical complication.
Free cement fragments were a different matter. Six of the 18 patients with FCF required additional surgery because of persistent pain, impaired mobility, or joint locking. Five of these underwent arthroscopic removal of the fragments, while one needed open surgery through the original incision after a fragment lodged in the posterior articular cavity, where arthroscopic access was not feasible. In that patient, intraoperative exploration revealed no obvious wear of the prosthesis or the polyethylene insert, so the operation concluded after the loose fragments were removed. All six patients who underwent surgical removal reported relief of knee pain and significant improvement in range of motion at follow-up.
The timing of symptoms varied widely. Among the six patients who needed reoperation, the interval between the primary surgery and the onset of symptoms ranged from 3 to 48 months. The authors speculate that repeated mechanical loading may gradually cause extruded cement to disintegrate and form free fragments within the joint, which can then impede flexion and extension or even lock the joint. They stress, however, that this proposed progression from extrusion to fragmentation remains a hypothesis, because the cross-sectional design of the study cannot establish how residual cement behaves over time.
Logistic regression comparing the 84 patients with residual cement against the remaining 237 patients identified one significant risk factor: the thickness of the polyethylene insert. Patients with thinner inserts, classified as 3 or 4 millimetres, had roughly 2.6 times the odds of residual cement, with an odds ratio of 2.59 and a 95 percent confidence interval of 1.38 to 4.98, p equal to 0.005. Age, body mass index, and the size of the implanted prosthesis did not differ significantly between the groups. A thinner spacer means a narrower working space between the femur and tibia, which the authors suggest makes complete cement clearance harder, particularly behind the tibial component.
The surgical technique itself may help explain why the problem arises at all. All operations were performed by experienced surgeons from a single team using Oxford mobile unicompartmental prostheses, with cement mixed from a single 40-gram pack in a closed vacuum system. Surgeons attempted to remove all visible cement, palpating the posterior margin of the tibial component with a curved dissector and irrigating the joint at the end of the procedure, yet residual cement still occurred in about a quarter of cases. The authors attribute this to the limited view of the posterior compartment in a minimally invasive partial knee approach, and to the fact that the anterior cruciate ligament and the femoral condyle restrict the angle at which the tibial component can be inserted, generating backward pressure that pushes cement posteriorly. In total knee replacement, by contrast, removal of the ligament allows vertical insertion of the tibial component and more uniform cement extrusion in all directions.
The findings align with earlier work. Hauptmann and colleagues previously reported a combined incidence of residual cement extrusion and free cement fragmentation of 20.8 percent, or 25 of 120 patients, while a cadaveric study by Scheele and colleagues detected free cement fragments in 66.7 percent of UKA specimens regardless of whether cemented or hybrid fixation was used. That high rate leads the authors to suggest that small free fragments after surgery may be a rather normal state, most of which never cause symptoms. Earlier literature had treated residual cement as a rare complication, but the authors note that most reports were individual case studies and that systematic data were lacking.
Management, the authors emphasise, should be stratified by symptoms rather than by the mere presence of cement. Asymptomatic patients or those with mild discomfort can be observed or treated conservatively. When symptoms persist despite adequate conservative therapy, accessible free fragments in the joint space can be removed arthroscopically, while fragments in the posterior cavity require open removal through the original incision. If the polyethylene insert shows wear, it must be replaced. The authors recommend intensified postoperative imaging follow-up for patients receiving thinner inserts, greater intraoperative attention to cement removal behind the tibial component, and reserving insert exchange or revision surgery for clearly defined indications such as evident polyethylene wear, aseptic loosening, or periprosthetic joint infection.
The study has important limitations that temper its conclusions. It was retrospective and lacked a predefined protocol, which may have introduced selection bias, and the sample size limited statistical power. Because CT was reserved for symptomatic patients while asymptomatic patients underwent plain radiography alone, the true prevalence of residual cement is probably higher than the reported 26.16 percent. The authors did not evaluate knee alignment morphotypes, which might confound the relationship between residual cement and mechanical complications, and the diagnosis rested on imaging in patients who mostly had no symptoms. Longitudinal evidence on whether residual cement inevitably progresses to symptoms, and on long-term prosthetic survival in affected patients, remains lacking, and the authors call for extended follow-up to clarify the natural history of the condition.
Subject of Research: Prevalence, risk factors and management of residual bone cement after unicompartmental knee arthroplasty
Article Title: Overlooked and underestimated: the prevalence and impact of mechanical cement-related complications in unicompartmental knee arthroplasty
Article References: Yu, Z., Wu, B., Lin, Y., Chen, R., Bai, G., Fang, X., Zhang, W., & Huang, Z. (2026). Overlooked and underestimated: the prevalence and impact of mechanical cement-related complications in unicompartmental knee arthroplasty. Arthroplasty, 8(1), Article 78. https://doi.org/10.1186/s42836-026-00439-x
Image Credits: AI Generated
DOI: 10.1186/s42836-026-00439-x
Keywords: unicompartmental knee arthroplasty, bone cement, residual cement extrusion, free cement fragmentation, polyethylene insert, knee osteoarthritis, arthroscopic removal, retrospective study, surgical complications, joint locking, risk factors, Overlooked
News Source: Ophelia Keating. (October 28, 2026). Residual bone cement after partial knee replacement is far more common than thought. Scienmag.
