Objective: The sliding hip screw (SHS) is the gold standard for the management of stable intertrochanteric (IT) fractures. However, intramedullary implants are now being increasingly used for management of unstable IT fractures especially those with a compromised or vulnerable lateral wall. Therefore, accurate classification of fracture is important to ensure proper surgical planning and choice of implant. The AO classification for IT fractures is based on plain radiographs alone and many authors have reported it to have poor inter- and intra-observer agreement. Therefore, the objectives of the study were to assess the improvement in inter- and intra-observer agreement of the AO classification after addition of CT scan to plain radiographs, to assess the change in pattern of AO classification on addition of CT scan to plain radiographs and to assess percentage of times, stable lateral wall seen on plain radiographs is classified as unstable or broken on CT scans.
Method: Fifty-four patients of intertrochanteric fracture were included in study. Plain radiographs of patients were shown to three orthopedic surgeons. They were asked to document the AO classification of the fracture, and comment on the integrity of the lateral wall. Then, CT scans with 3D reconstructions of the same patients were provided along with the radiographs and they were asked to classify the fracture again. Inter- and intra-observer agreement of the AO classification based on plain radiographs alone and once CT scan with 3D reconstruction was added to the plain radiographs was determined using Kappa coefficients. Pattern of change in classification on addition of CT scan to plain radiographs was also assessed.
Result: The mean kappa value for inter- and intra-observer agreement for AO classification on plain radiographs alone were 0.58 (moderate) and 0.66 (substantial), respectively. Upon addition of CT scan to plain radiographs, both improved to 0.70 (substantial) and 0.77 (substantial), respectively. The AO classification of the fracture changed 28.70% times (93 of 324 observations) upon addition of CT scan to plain radiographs. 96.77% times (90 of 93 observations) the classification was upgraded to higher group, while it was downgraded in only 3.22% times (3 of 93). 55.91% times this change was observed in AO 31 A 2 group (52 of 93 observations). In 17.59% cases (57 of 324 observations), fractures which were classified as stable (A1.1-A2.1) on radiographs alone, were reclassified as unstable (A2.2-3.3) upon addition of CT scan to plain radiographs. In 11.4% cases (37 of 324 observations), the lateral wall was classified as vulnerable or broken on CT scans where it was classified intact on plain radiographs.
Conclusion: Addition of CT scans with 3D reconstructions to plain radiographs improves the intra- and inter-observer agreement of the AO classification. Addition of CT scan results in change in classification of the fracture in about one out of three cases. This most commonly happens in the AO 31 A 2 group. Most of the times, this results in the classification of fracture being upgraded. Many fractures which are initially classified as stable (A1.1-A2.1) on radiographs are reclassified as unstable (A2.2-3.3) on addition of CT scans. The lateral wall is also classified as vulnerable or broken more number of times on CT scans than plain radiographs alone. Therefore, we conclude that CT scan with 3 D reconstructions definitely helps in better pre-operative classification of intertrochanteric fractures especially in select group of fractures (AO 31 A 2) where stability and integrity of lateral wall is difficult to assess.
Keywords: AO classification; CT scan; Intertrochanteric fractures.
© Indian Orthopaedics Association 2021.