Abstract
Digital radiography has replaced traditional radiography in many hospitals yet little is known regarding the accuracy of this new technology in THA templating. Our study analyzed the reproducibility and reliability of computer templating in primary uncemented THA as compared to standard on-lay templating techniques with hardcopy radiographs from a digital source.
In December 2004 our hospital converted from standard hardcopy radiography to digital radiography. Patients undergoing THA had preoperative digital radiographs taken which included an AP pelvis with a 50mm magnification marker place in the groin, AP hip, and a cross-table lateral. Forty patients were selected that met our inclusion criteria to begin the templating process, including adequate placement of the magnification marker and optimal hip implant positioning on the postoperative films. Cases that did not have a marker, had significant deformity or mal-positioned implants were excluded. Hardcopies of digitised radiographs were printed and a traditional templating technique using 120% magnified on-lay transparent templates was performed (Capello, 1984). Digital templating was performed using OrthoView Software (Bono, 2004). Templating was conducted by two staff surgeons and one resident. Intra-observer and inter-observer effects were calculated using an Intraclass Correlation Coefficient (ICC).
Digital templating showed good inter-observer and intra-oberserver reliability with ICC values > 0.7. Using computer templating, the femoral stem was accurately predicted to within one size in 85% of the cases, and the acetabular cup was accurately predicted to within one size in 80% of the cases.Using on-lay templating, the femoral stem was accurately predicted to within one size in 85% of the cases, and the acetabular cup was accurately predicted to within one size in 60% of the cases.
There was a significant difference between the accuracy of acetabular cup templating between techniques, likely as a result of the ability of computer templating to adjust for magnification error. Computer templating was able to accurately predict postoperative leg-length discrepancy, abduction angle and horizontal offset. Overall, computer templating was found to be a reproducible and reliable technique for uncemented THA. Problems and lessons learned in the implementation of a digital templating system will be discussed.
Correspondence should be addressed to: Cynthia Vezina, Communications Manager, COA, 4150-360 Ste. Catherine St. West, Westmount, QC H3Z 2Y5, Canada
