United States – October 27, 2026 -- Robotic assistance in total knee arthroplasty (TKA) rose from 12.4% of procedures in 2018 to 36.1% in 2026, a 191% relative increase, according to an analysis of 362,667 procedures in Truveta Data.
Robotic assistance was used in 24.3% of all procedures in the study period, which ran from January 2018 to August 2026. Robotic-assisted cases ran longer. Median operation time was 136 minutes (IQR 116-160), against 126 minutes (IQR 109-149) without robotics.
Device documentation lifts robotic case identification to 22.2% from 14.3% by coding alone
The study identified robotic cases through procedure add-on codes or documentation of robotic device use. Device documentation flagged 22.2% of procedures as robotic-assisted, while procedure coding alone flagged 14.3%. Prior analyses often relied on procedure codes alone. The authors say estimates based on coding alone match earlier published utilization figures, and that adding device documentation captures a substantially larger share of robotic cases.
Robotic cases show lower transfusion and 90-day complication rates
Blood transfusion within two days of surgery occurred in about 0.2% of robotic-assisted procedures versus 0.6% without robotics. Within 90 days, robotic-assisted TKA showed lower rates of:
- Infection: 1.6% vs. 2.5%
- Revision surgery: 0.3% vs. 0.5%
- Venous thromboembolism: 1.9% vs. 2.2%
- Wound dehiscence: 0.7% vs. 0.8%
The authors describe the absolute wound dehiscence gap as small, although it remained significant after adjustment.
One-year revision rate is 0.9% for robotic cases versus 1.3% for conventional
Between 90 days and one year, robotic-assisted procedures had lower rates of revision surgery (0.9% vs. 1.3%) and prosthetic loosening (0.2% vs. 0.4%). Revision surgery, loosening and wear or osteolysis (0.1% in both groups) remained significantly less likely after adjustment, despite the low absolute rates.
Stiff knee diagnoses ran against the trend, at 3.5% for robotic cases versus 3.0%, and the difference stayed significant after adjustment. Pain-related diagnoses were similar between groups at 24.4% and 23.7%.
Adjusted models control for age, sex, obesity and comorbidity burden
Multivariable logistic regression adjusted for sex, age, race, ethnicity, rural versus urban residence, obesity, year of surgery and Elixhauser Comorbidity Index. The cohort had a median age of 69 years. Women made up 60.1% of patients, 69.5% lived in urban areas and 74.8% had documented obesity.
The authors note that robotic systems carry higher costs and that evidence on short-term complications remains incomplete. Prior real-world studies reported lower venous thromboembolism and transfusion rates after robotic TKA but less consistent findings on infection.