The reported incidence of luxation as a complication following THR has ranged from 1.3%1 to 15%2. Potential risk factors described in the literature have included implant orientation, stem subsidence, soft tissue tension and implant impingement2,3,4,5,6,7,8. Reported outcomes for surgical revision of luxation are limited, though success rates have ranged from 0-80%7,9,10,11,12,13,14. Pelvic osteotomies have been used to address luxations attributed to an excessively open cup1,9,10 with mixed outcomes. However, as the axis of rotation at the osteotomy site is not aligned with the axis about which the cup is opened or closed, pelvic osteotomies affect cup version as well as lateral opening. Other reported techniques include neck lengthening, increasing cup and head diameter, iliofemoral suture placement and ventral acetabular augmentation7,8,12.
Studies looking at risk factors for postoperative luxation have used radiographs to assess implant orientation. The Biomedtrix BFX cup is effectively a truncated hemisphere with a larger open face at the equator of the hemisphere and a smaller truncated face produced by removal of a portion of the hemisphere to match the normal anatomy of the acetabulum. Renwick et al.15 proposed a method to assess BFX cup version in which a line is drawn across the junction of the truncated face with the open face on the ventrodorsal hip radiograph. This allows measurement of the version of the truncated face, though not of the open face. It can be argued that while both truncated face and open face version are important aspects of cup position open face version is more clinically relevant since the relationship between the open face and the prosthetic neck during internal and external rotation of the limb will determine the impingement-free range of movement. We have shown that truncated face version is not a reliable indicator of open face version and that normal truncated face version can mask excessive open face retroversion when there is excessive declination of the cup16. Furthermore, angles of version and the angle of lateral opening (ALO) are dependent on the image planes in which they are being measured. Consequently, computed tomography (CT) has been shown in people to outperform radiography for assessment of acetabular cup orientation17,18. When using 2-dimensional imaging, the effects of inclination, ALO and version are interlinked. This may explain the conflicting evidence in the literature regarding the influence of ALO on luxation risk2,3 and the lack of evidence regarding the influence of cup version.
Due to concerns regarding our ability to accurately assess implant orientation radiographically, we routinely use CT for postoperative imaging and have established a protocol to assess stem anteversion and cup inclination, ALO and version as independent variables. In a dog with a postoperative luxation, we pay particular attention to open face version of the cup, ALO and the match between the degree of cup retroversion and stem anteversion. If any of these values exceed what we would consider to be a safe range, we will plan to revise implant positioning, most commonly of the cup, though femoral anteversion may be adjusted in some cases.
We generally see luxations within the first few weeks of surgery and as we predominantly use BFX implants, extraction is usually relatively straightforward. At surgery we reduce the hip, assess soft tissue tension and carefully assess range of motion, aiming to reproduce the mechanism of luxation. We have found that this most commonly occurs on hyperflexion and internal rotation of the hip, with luxation on external rotation of the hip being less common and luxation on adduction or abduction of the femur being very rare. Luxation on hyperflexion and internal rotation of the hip is usually due to contact between the neck and the margin of the cup, though there can be impingement between the proximal femur and osteophytes or soft tissue hypertrophy at the cranial aspect of the acetabulum. Cross et al.2 felt that the final resting point of the femur could not be taken as an indicator of the direction of luxation and our experience has been similar with cranial impingement potentially resulting in caudoventral or dorsal luxation in different dogs.
An increased risk of luxation has been identified in dogs with luxoid hips and with low femoral displacement postoperatively and it has been suggested that soft tissue laxity may be the predisposing factor5,6,7. These dogs typically have high soft tissue tension at reduction resulting in placement of a +0 head. While subsequent tissue relaxation may play a role, we suspect that the proximity between the proximal femur and the pelvis inherent with short neck length is more likely to cause bony or soft tissue impingement at the extremes of range of motion and that the effectiveness of neck lengthening in these dogs is due to the reduction in impingement as well as restoration of soft tissue tension. If impingement with periacetabular osteophytes and soft tissue hypertrophy is identified, then this tissue is resected during revision.
The combination of a small size head (12,13 or 14mm) and a standard stem results in a reduced ratio between the diameters of the head and neck and a corresponding reduction in the range of motion. If luxation occurs despite appropriate implant orientation, then the diameter of the cup may be increased to allow a 17mm head even if the cup appears oversized. When removing and replacing a cup, we insert a 2mm IM pin into the ilium cranial to the acetabulum and aligned with the axis of the cup impactor which is placed. This provides a reference axis against which we can adjust cup orientation.
Our experience is that surgical revision following postoperative luxation is generally successful provided the mechanism of luxation can be identified and addressed. There remain, however, a small group of dogs in which a clear underlying cause is not found and in which abnormal neuromuscular control may play a role.
References