A large-scale nationwide cohort study from Japan has demonstrated that the perioperative safety advantages of unicompartmental knee arthroplasty compared with total knee arthroplasty depend strongly on patient age. Evaluating records from hundreds of thousands of surgical admissions across Japan, researchers observed that unicompartmental knee arthroplasty significantly decreased the incidence of major systemic complications requiring additional medical or surgical intervention among individuals aged 79 years or younger. In sharp contrast, this protective advantage against major systemic complications disappeared among patients aged 80 years and older, for whom the incidence of severe postoperative systemic events did not differ significantly between the two surgical approaches. Postoperative red blood cell transfusion requirements, however, were consistently and substantially lower following unicompartmental knee arthroplasty across all evaluated age groups, providing clinicians with precise, age-stratified evidence to guide individualized surgical decision-making for patients with degenerative joint disease.
Total knee arthroplasty has long served as the established, standard surgical intervention for extensive knee conditions including advanced osteoarthritis, osteonecrosis, and rheumatoid arthritis, particularly among middle-aged and older populations. Concurrently, unicompartmental knee arthroplasty has gained widespread utilization as a targeted surgical treatment for disease confined strictly to a single knee compartment. Although knee arthroplasty demand continues to rise steadily alongside aging and overweight populations, the optimal surgical strategy for minimizing systemic complications has remained an active clinical debate. Historically, unicompartmental knee arthroplasty has been recognized as a less invasive procedure that preserves native bone and soft tissue structures, minimizes intraoperative blood loss, accelerates functional recovery, and reduces initial hospital costs, despite carrying known disadvantages such as higher rates of periprosthetic fracture and long-term implant revision compared with total replacement. Prior research suggested lower general complication rates following partial knee procedures, yet whether these systemic protective benefits persist equally across advanced age categories had remained unresolved due to a lack of adequately powered comparative studies.
To overcome the statistical limitations of smaller institutional cohorts when assessing rare but critical postoperative events, the research team conducted a retrospective analysis using the Japanese Diagnosis Procedure Combination database. This administrative inpatient database encompasses discharge abstracts and hospital claims from academic and acute care institutions across Japan, comprising data from 1,171 participating hospitals and covering approximately 7.16 million patient admissions, which represents roughly half of all acute care hospitalizations nationwide. Participation in the database is mandatory for all Japanese university and academic hospitals and voluntary for community medical centers, capturing surgical records where diagnoses and interventional procedures are systematically coded by attending physicians according to the International Classification of Diseases, Tenth Revision. Previous validation research had verified high diagnostic coding accuracy within this administrative claims repository, establishing a robust foundation to investigate uncommon perioperative outcomes associated with elective orthopedic interventions.
The study population was selected from 398,036 patients who underwent knee arthroplasty within 14 days of hospital admission between July 2010 and March 2022. The investigators applied rigorous exclusion criteria: patients receiving bilateral knee arthroplasty during a single hospital admission, those undergoing non-standard or unspecified knee procedures, and individuals whose primary diagnosis was not osteoarthritis, osteonecrosis, or rheumatoid arthritis were excluded from analysis. Following the removal of 39,377 cases that met the exclusion criteria, the final analyzed cohort comprised 358,659 surgical admissions, consisting of 36,235 patients who underwent unicompartmental knee arthroplasty and 322,424 patients who received total knee arthroplasty. Prior to statistical adjustment, notable differences in baseline characteristics were evident between the groups, including imbalances across patient sex, body mass index categories, smoking status, underlying primary diagnoses, preoperative corticosteroid use, immunosuppressant administration, and selected anesthesia modalities. Hospital stay durations also reflected typical Japanese clinical practice, with a median postoperative length of stay of 19 days in the unicompartmental arthroplasty cohort compared with 23 days in the total knee arthroplasty cohort.
To ensure diagnostic specificity and capture severe clinical events following elective joint surgery, the investigators established a primary composite outcome consisting of postoperative in-hospital death and major systemic complications that necessitated additional therapeutic procedures or invasive interventions during the surgical admission. These specified complications comprised cardiac arrest, acute coronary events, heart failure, aortic aneurysm or dissection, respiratory failure requiring mechanical ventilation, pulmonary embolism, cerebrovascular events, acute renal failure requiring hemodialysis, and gastrointestinal bleeding or peptic ulcer perforation. Secondary endpoints included each individual component of the composite outcome alongside the requirement for postoperative red blood cell transfusion. To control for confounding indications and baseline discrepancies between surgical cohorts without sacrificing statistical power, the researchers calculated propensity scores via logistic regression and implemented stabilized inverse probability of treatment weighting. Confounders adjusted through weighting included age categories, sex, body mass index classifications, smoking status, primary diagnosis, preexisting systemic comorbidities, steroid and immunosuppressant use, and anesthesia type, achieving excellent post-weighting covariate balance with standardized differences below 0.1.
Across the entire adjusted study population, unicompartmental knee arthroplasty demonstrated an overall protective association against severe in-hospital events compared with total knee arthroplasty. After stabilized inverse probability of treatment weighting adjustment, the incidence of the primary composite outcome was significantly lower in the unicompartmental arthroplasty cohort, yielding a risk ratio of 0.65 with a 95 percent confidence interval ranging from 0.50 to 0.85 and a p-value of 0.001. When evaluating specific complications across the full cohort, pulmonary embolism occurred significantly less often after unicompartmental procedures, showing a risk ratio of 0.53 with a 95 percent confidence interval of 0.29 to 0.94 and a p-value of 0.031. Furthermore, the postoperative red blood cell transfusion rate was drastically decreased among unicompartmental knee arthroplasty patients, demonstrating a risk ratio of 0.09 with a 95 percent confidence interval of 0.08 to 0.11 and a p-value below 0.001. Initial multivariable logistic regression across 334,704 patients similarly identified an adjusted risk ratio of 0.57 for unicompartmental surgery, while revealing that complication risks escalated markedly with advancing chronological age, with odds ratios reaching 1.32 for patients aged 80 to 89 years and 2.05 for those aged 90 years or older compared with patients aged 70 to 79 years.
Subgroup analyses stratified by patient age demonstrated that the relative clinical benefits of unicompartmental knee arthroplasty differed substantially between older and younger surgical populations. Among patients aged 79 years or younger, unicompartmental knee arthroplasty was associated with a marked 52 percent reduction in the primary composite outcome compared with total knee arthroplasty, producing an adjusted risk ratio of 0.48 with a 95 percent confidence interval of 0.34 to 0.68 and a p-value below 0.001. An examination of specific major complications within this younger cohort revealed statistically significant risk reductions favoring unicompartmental surgery across multiple organ systems. Specifically, unicompartmental arthroplasty was associated with significantly decreased risks of postoperative heart failure with a risk ratio of 0.37, respiratory failure requiring mechanical ventilation with a risk ratio of 0.47, pulmonary embolism with a risk ratio of 0.49, and red blood cell transfusion with a risk ratio of 0.07. In-hospital mortality was also lower following unicompartmental surgery among patients aged 79 years or younger with a risk ratio of 0.36, although this difference fell just short of conventional statistical significance with a 95 percent confidence interval of 0.12 to 1.06 and a p-value of 0.065.
The clinical picture changed dramatically when the investigators examined the cohort of patients aged 80 years and older. Within this elderly demographic, unicompartmental knee arthroplasty provided no detectable protective advantage against severe systemic adverse events compared with total knee arthroplasty. Following stabilized inverse probability of treatment weighting adjustment, the incidence of the primary composite outcome among patients aged 80 years and older yielded a risk ratio of 0.92 with a 95 percent confidence interval spanning 0.62 to 1.35 and a non-significant p-value of 0.670. Likewise, unadjusted comparisons and adjusted evaluations of individual systemic endpoints, including cardiac events, acute respiratory failure, pulmonary embolism, cerebrovascular accidents, acute renal failure, and in-hospital death, showed no statistically significant differences between unicompartmental and total knee arthroplasty in this octogenarian and older group. The sole clinical endpoint that consistently favored unicompartmental knee arthroplasty among patients aged 80 years and older was postoperative red blood cell transfusion, which remained significantly reduced with an adjusted risk ratio of 0.12, a 95 percent confidence interval of 0.10 to 0.14, and a p-value below 0.001.
Interpreting these findings, the study authors suggested that the physiological advantages typically conferred by the less invasive nature of unicompartmental knee arthroplasty may become attenuated by heightened baseline perioperative vulnerability and systemic frailty in very elderly individuals. While younger patients appear able to translate reduced surgical tissue trauma, preserved bone stock, and limited blood loss into lower systemic organ stress and fewer life-threatening cardiopulmonary complications, the diminished physiological reserve of patients aged 80 years and older may render them equally susceptible to major systemic decompensation regardless of the specific surgical approach employed. The authors noted that this observation aligns with broader surgical literature documenting elevated perioperative morbidity and mortality rates among octogenarians undergoing major operations across various medical specialties. Previous smaller investigations, such as a single-center study comparing 30 unicompartmental cases against 90 total knee replacements in patients aged 85 years and older that reported fewer complications after partial replacement, were constrained by limited sample sizes and age-matching methodologies that lacked adequate statistical power to control for broader confounding factors.
These findings provide essential nuance for orthopedic surgeons counseling elderly candidates regarding elective knee reconstruction. The authors emphasized that the absence of a significant difference in major systemic complications among octogenarians does not mean that unicompartmental knee arthroplasty lacks clinical utility in this demographic. Rather, unicompartmental arthroplasty continues to confer meaningful advantages in elderly patients, including consistently lower blood transfusion requirements and established benefits documented in prior literature, such as accelerated postoperative rehabilitation, superior functional joint scores, and earlier mobility. However, surgeons must not assume that choosing a partial knee replacement inherently shields patients aged 80 years and older from catastrophic cardiopulmonary, thromboembolic, or cerebrovascular events. Because the risk of in-hospital mortality and major systemic complications requiring additional medical or surgical procedures remains comparable to total knee arthroplasty in this elderly group, careful and rigorous perioperative risk assessment is equally critical whenever unicompartmental arthroplasty is considered for very elderly patients.
The investigators also detailed several inherent methodological limitations of their nationwide administrative study that warrant consideration when interpreting the results. First, the Diagnosis Procedure Combination database lacks granular clinical indicators, such as exact anatomical disease severity, radiological grading, and surgeon technical expertise or hospital procedural volume, which have been documented to influence arthroplasty outcomes. Although propensity score weighting achieved excellent balance across recorded baseline variables, residual confounding related to patient selection and surgical indication cannot be entirely excluded. Second, administrative claims data carry a potential risk of coding misclassification, though past validation studies demonstrated high sensitivity and specificity for surgical diagnoses within the Japanese registry. Third, the primary endpoint focused strictly on in-hospital death and major systemic complications requiring active medical or surgical interventions, meaning minor complications resolving without procedural intervention were not captured. Finally, because the database exclusively tracks inpatient admissions and excludes outpatient procedures, post-discharge complications and long-term outcomes, including prosthesis revision rates and overall implant survivorship, could not be evaluated, although the comparatively long inpatient hospitalizations in Japan ensured comprehensive capture of early postoperative adverse events.
Subject of Research: Age-stratified comparison of major systemic complications and blood transfusion rates between unicompartmental and total knee arthroplasty
Article Title: Age-stratified risk of major systemic complications following unicompartmental versus total knee arthroplasty: a nationwide cohort study
Article References: Kawata, M., Aso, S., Kawaguchi, K., Yamagami, R., Kono, K., Taketomi, S., Matsui, H., Fushimi, K., Yasunaga, H., & Tanaka, S. (2026). Age-stratified risk of major systemic complications following unicompartmental versus total knee arthroplasty: a nationwide cohort study. Knee Surgery & Related Research, 38(1), Article 35. https://doi.org/10.1186/s43019-026-00330-8
Image Credits: AI Generated
DOI: 10.1186/s43019-026-00330-8
Keywords: unicompartmental knee arthroplasty, total knee arthroplasty, knee replacement surgery, major systemic complications, perioperative risk, age stratification, geriatric surgery, in-hospital mortality, red blood cell transfusion, pulmonary embolism, propensity score weighting, Diagnosis Procedure Combination, orthopedic surgery, postoperative outcomes, surgical invasiveness, clinical epidemiology
Cite Scienmag News
Ophelia Keating. (November 18, 2026). Partial Knee Replacement Lowers Systemic Complications Only in Patients Under Eighty. Scienmag. https://scienmag.com/partial-knee-replacement-lowers-systemic-complications-only-in-patients-under-eighty/
Ophelia Keating. "Partial Knee Replacement Lowers Systemic Complications Only in Patients Under Eighty." Scienmag, 18 November 2026, https://scienmag.com/partial-knee-replacement-lowers-systemic-complications-only-in-patients-under-eighty/. Accessed 18 November 2026.
Ophelia Keating. "Partial Knee Replacement Lowers Systemic Complications Only in Patients Under Eighty." Scienmag. November 18, 2026. https://scienmag.com/partial-knee-replacement-lowers-systemic-complications-only-in-patients-under-eighty/








