Fig. 1 Knee range of motion at three months, one year, and two years
Courtesy of Nana Sarpong, MD


Published 9/2/2021
Rebecca Araujo

Sensor-guided Ligament Balancing in TKA Does Not Improve Short-term Outcomes versus Freehand Technique

In a randomized, controlled trial, sensor-guided ligament balancing during total knee arthroplasty (TKA) was found to be noninferior to freehand balancing but did not lead to improvements in short-term postoperative function or outcomes. The findings will be presented today by Nana Sarpong, MD, an adult reconstruction and joint replacement surgery fellow at Hospital for Special Surgery.

Soft-tissue imbalance accounts for more than one-third of early TKA revisions, according to the study authors. Rebalancing, which is typically done freehand by orthopaedic surgeons, is necessary to realign the lower extremity to a neutral mechanical axis during surgery. Dr. Sarpong and colleagues at Irving Medical Center sought to determine whether using a sensor-guided balancing device would lead to enhanced TKA outcomes compared to standard care.

“Soft-tissue balancing in total knee replacements has traditionally been more of an art than a science, relying primarily on the surgeon’s subjective assessment based on nebulous tactile feedback,” Dr. Sarpong told AAOS Now Daily Edition. “Sensor-guided technology helps to quantify the soft-tissue balance in real time and could improve balance. Improved balance may contribute to decreased pain, improvement in function, and increased patient satisfaction.”

For their study, the investigators prospectively enrolled 130 patients who underwent primary TKA at Columbia University Irving Medical Center beginning in December 2017. Sixty-three patients were randomized to sensor-guided soft-tissue balancing (S), and 67 patients were assigned to freehand balancing (NS). The planned follow-up is a minimum of two years, but the present analysis assessed outcomes at three months postoperatively.

Study authors assessed the following variables: patient-reported outcomes (PROs), knee range of motion (ROM), visual analog scale (VAS) for pain, opioid consumption, inpatient physical therapy performance, length of stay (LOS), discharge disposition, and incidence of arthrofibrosis. The noninferiority threshold was established as a 5-degree ROM difference, with α = 0.05 and P = 0.80. A chi-square test was used to assess categorical variables.

At three months, knee ROM was comparable between treatment groups. Average ROM was 115.7 ± 11.4 degrees for the S cohort, compared to 114.9 ± 12.5 degrees in the NS cohort (P = 0.76), which met the noninferiority threshold. The authors noted a nonsignificant increase in operative time associated with sensor-guided balancing compared to freehand balancing.

The authors also reported a statistically significant improvement in PROs for the NS cohort compared to the S cohort. Western Ontario and McMaster Universities Arthritis Index (WOMAC) scores for pain and physical function favored freehand versus sensor-guided balancing. Scores for the 12-item Short Form Survey–Pain and Knee Society Function Score–Physical Function were also higher in the NS cohort than in the S cohort. However, the authors noted that none of the differences reached the threshold for minimal clinical importance.

There was no significant difference between groups in operative time, LOS, physical therapy performance, VAS pain scores, opioid use, discharge disposition, arthrofibrosis, or rates of complications or reoperation.

Dr. Sarpong concluded: “In the hands of experienced arthroplasty surgeons, utilization of a sensor-balancing device for soft-tissue balancing in TKA increases operative time but does not result in clinically perceived differences in knee ROM, PROs, and secondary outcomes.”

The findings do not mean that sensor-guided ligament balancing does not have a place in practice. In fact, he said, “Inexperienced surgeons may benefit from this technology.”

Study investigators are continuing to collect long-term outcomes and to monitor patients for other clinically significant differences between the study groups. At one and two years, average knee ROM for the sensor-guided group showed improvement to 118.8 degrees and 116.2 degrees, respectively, compared to 116.5 degrees and 116.9 degrees, respectively, for the freehand group (Fig. 1; P = 0.41 and P = 0.87).

“In the future, we hope to assess soft-tissue balance dynamically through a full knee arc of motion, instead of just 10, 45, and 90 degrees of flexion,” added Dr. Sarpong.

The study was limited by a lack of standardization in implant selection among the three arthroplasty surgeons participating in the study, which “may have affected our findings,” noted Dr. Sarpong. He added that, “This [limitation] is reconciled as it allows for greater external validity and prevents the effects of the surgeons using unfamiliar implant systems.”

The study will be presented as Paper 450 today at 8 a.m. in Room 3.

Dr. Sarpong’s coauthors of “A Randomized, Controlled Trial of Outcomes in Freehand versus Sensor-guided Balancing in Total Knee Arthroplasty: Results from the Short Term” are Matthew Grosso, MD; Michael B. Held, MD; Carl L. Herndon, MD; Fnu Akshay, MD; Roshan P. Shah, MD; Herbert John Cooper, MD; and Jeffrey A. Geller, MD.

Rebecca Araujo is the associate editor of AAOS Now. She can be reached at