[Total arthroplasty, using Hardinge's approach, combined with trochanterotomy: comparative results of 200 cases]

A retrospective study to compare the results obtained in our first 100 total hip prostheses inserted by Hardinge's approach and of the 100 others inserted by trochanterotomy was undertaken.

In the trochanterotomy group the average age was 66 years; average follow-up was 26 months. Using the Merle d'Aubigné score the initial score was 11.8. There were 65 cases of centered hip arthritis. In the Hardinge group the average age was 65 years; average follow-up was 28.3 months. The Merle d'Aubigné initial score was 12.3. There were 78 cases of centered hip arthritis. There were therefore no significant differences between the two groups and the two groups were comparable.

The quantitative variables (age, duration of operation, blood loss, blood transfusion, follow-up) were compared by Student's test. The qualitative variables (thrombo-embolic complications, dislocations, periarticular ossifications, acetabular radiolucency lines, non-union of the greater trochanter, gluteus medius palsies) were compared by the chi 2 test.

We found no significative differences on neither the functional level nor on the orientation of the prostheses nor on the number of infectious complications between these two surgical approaches. Moreover, we found more complications such as thromboembolism and dislocations favoured by non-union of the greater trochanter in patients operated by trochanterotomy. These patients also had greater blood loss. In patients operated by Hardinge's approach, we found gluteus medius palsies (recovering secondarily); we also found a higher frequency of periarticular ossifications and a greater number of partial acetabular lines.

Non-union of the greater trochanter appears in all the series of total hip arthroplasty by trochanterotomy. No technique permitted to avoid this complication which usually leads to pain and hip instability. This surgical approach is associated with higher blood loss. With Hardinge's approach there is no risk of non-union of the greater trochanter and blood loss is less important. The risk of gluteus medius palsy has to be taken in to account but digital dissection of the muscle fibers seems adequate to diminish the frequency of this complication. There is also a greater number of asymptomatic periarticular ossifications in our study but whose long term consequences are unknown.

This study leads us to prefer the Hardinge approach for total hip arthroplasty. Our recent experience encourages us even because it permits osteoplastic ridge and total hip resumption. We use the trochanterotomy only for the most difficult cases specially hip arthritis secondary to severe dysplasia or congenital hip dislocations when a lowering effect of the great trochanter should also be associated.