Kristen Thompson1, Will Griffiths-Jones2, Lisa Frendin1, Jil Wood1, Ian A Harris3, Darren B Chen1, Samuel J MacDessi4. 1. Sydney Knee Specialists, Kogarah, NSW, Australia. 2. North Devon District Hospital, Raleigh Park, Barnstaple, United Kingdom. 3. Ingham Institute for Applied Medical Research, South Western Sydney Clinical School, University of New South Wales, Sydney, NSW, Australia. 4. Sydney Knee Specialists, Kogarah, NSW, Australia; St George Hospital Clinical School, University of New South Wales, Sydney, NSW, Australia. Electronic address: samuelmacdessi@sydneyknee.com.au.
Abstract
BACKGROUND: Sensor-guided compartmental pressure measurements are becoming increasingly utilized in total knee arthroplasty (TKA) to objectively confirm intraoperative knee balance. We aimed to determine agreement of pressure measurements between two observers when performing sensor-guided TKA with the use of computer-assisted surgery (CAS). METHODS: One-hundred and eighteen consecutive patients undergoing 130 TKAs were analysed. Femoral and tibial trial implants were inserted prior to performing knee balancing. We compared the reliability of sensor pressure compartmental measurements between two observers at 10, 45 and 90° of flexion using intraclass correlation coefficient (ICC) estimates and the 95% limits of agreement (Bland-Altman plots). RESULTS: The interobserver agreement between sensor pressure measurements was excellent at 10° of knee flexion, with ICCs of 0.93 and 0.91 in the medial and lateral compartments, respectively (P < 0.001). At 45°, medial and lateral compartment ICCs were 0.91 and 0.76, respectively (P < 0.001). At 90°, the ICC was 0.88 medially and 0.76 laterally (P < 0.001). Although the agreement decreased at higher knee flexion, it remained good to excellent. The 95% limits of agreement at each angle were all within 20 psi and 11 psi for the medial and lateral compartments, respectively. CONCLUSIONS: There was excellent interobserver agreement of sensor pressure measurements at 10° of knee flexion with computer-assisted TKA. Interobserver agreement decreased slightly as knee flexion angles increased, particularly in the lateral compartment. It is likely that interobserver agreement and hence reliability of sensor pressure measurements in TKA has some dependence on accurate angular positioning of the knee.
BACKGROUND: Sensor-guided compartmental pressure measurements are becoming increasingly utilized in total knee arthroplasty (TKA) to objectively confirm intraoperative knee balance. We aimed to determine agreement of pressure measurements between two observers when performing sensor-guided TKA with the use of computer-assisted surgery (CAS). METHODS: One-hundred and eighteen consecutive patients undergoing 130 TKAs were analysed. Femoral and tibial trial implants were inserted prior to performing knee balancing. We compared the reliability of sensor pressure compartmental measurements between two observers at 10, 45 and 90° of flexion using intraclass correlation coefficient (ICC) estimates and the 95% limits of agreement (Bland-Altman plots). RESULTS: The interobserver agreement between sensor pressure measurements was excellent at 10° of knee flexion, with ICCs of 0.93 and 0.91 in the medial and lateral compartments, respectively (P < 0.001). At 45°, medial and lateral compartment ICCs were 0.91 and 0.76, respectively (P < 0.001). At 90°, the ICC was 0.88 medially and 0.76 laterally (P < 0.001). Although the agreement decreased at higher knee flexion, it remained good to excellent. The 95% limits of agreement at each angle were all within 20 psi and 11 psi for the medial and lateral compartments, respectively. CONCLUSIONS: There was excellent interobserver agreement of sensor pressure measurements at 10° of knee flexion with computer-assisted TKA. Interobserver agreement decreased slightly as knee flexion angles increased, particularly in the lateral compartment. It is likely that interobserver agreement and hence reliability of sensor pressure measurements in TKA has some dependence on accurate angular positioning of the knee.