The present study took a unique approach by combining gait analysis, strength testing and novel MR imaging to examine the effect of surgical approach on post-operative hip function. The results suggest that patients who have undergone THA via a posterior approach, have slightly poorer gait function and muscle strength compared to a DLA group, particularly during fast walking. In spite of this the PROMs data indicates that the impact of this on quality of life is minimal, with good outcomes reported in both surgical groups.
A key finding in this study is that the PA group tended not to extend the hip, regardless of the increasing demands between walking conditions, whereas the DLA group demonstrated hip extension under fast walking, similar to control participants (see Table 3). These findings suggest that the PA group may have persisting functional imitations due to the iatrogenic effects of surgery, compared to DLA and control participants. Previous work has found similar hip extension deficits in post THA patients during walking, irrespective of surgical approach (6, 8). The apparent group difference in hip flexion-extension kinematics had a systematic effect on other gait measures. During walking, both surgical groups demonstrated reduced single support time (as a % of stance phase), walking velocity and hip extension power compared to the control group (Table 3). However, when required to increase walking velocity i.e., during the fast walking condition, the DLA group systematically increased single support time, walking velocity and stride length and hence loaded the limb similar to the control participants, whereas the PA group, were systematically slower, with shorter stride lengths and reduced vGRF impulse. To our knowledge, this is the first study to demonstrate gait differences between the PA and DLA surgical groups, potentially due to the novel inclusion of more demanding ADLs which are likely to exacerbate any differences. This is in accordance with previous studies in hip OA cases, which have shown that when required to walk at an increased velocity, symptomatic hip OA patients achieved a greater hip extension to adapt to the greater demand (27), and suggests that functional differences are accentuated when THA patients perform more demanding ADLs (15).
Despite the functional differences between the surgical groups, the PROMs would suggest that both surgical groups perceived hip function and physical health to be good following THA and indeed the post-operative scores were superior compared to the control group. These PROMs outcomes suggest that, regardless of approach and apparent minor functional differences, patients report good post-operative quality of life and hip function, a finding reported previously for both lateral and PA THA (4). The superior physical health reported by both surgical groups, compared to the control group in the SF-36 questionnaire can likely be attributed to a response shift due to the positive change in relative health state after surgery, conferring a more positive outlook in the post operative THA patients (28).
The PA group demonstrated generally weaker hip muscles compared to the DLA and control participants, potentially as a consequence of greater soft tissue disruption arising from their surgical approach. It is possible that the dissection of the Gluteus Maximus reduces extensor power, which in turn reduces the ability to extend the hip during walking/fast walking, thus suggesting a link between muscle strength and locomotor recovery (29). This has been further emphasised in a recent randomised control trial, where the DLA was found to produce superior functional recovery compared the PA (30). Previous research, which also found no difference in walking between PA and DLA groups, had reported greater hip abductor and flexor strength, in PAs compared to DLA group (31) and so future work should consider combining muscle strength measures and more demanding ADLs in order to understand in detail the role of THA and/or surgical approach on real-world physical function.
In the present study we took the unique approach of measuring muscle CSA in a small, subset of patients to examine the effect of surgical approach on muscle condition. Compared to the contralateral limb, the DLA group had a systematically reduced Gluteus Medius CSA in the operated limb, a prime abductor, compared to the PA group of the same muscle. Thus, whilst there was no between-group abductor strength differences, the MRI indicated that the iatrogenic effect on the CSA of the same muscle is greater in the DLA group. The reduced CSA relevant muscle in the operated limb in both surgical groups suggests that the musculature disrupted during surgery remains atrophied up to five years post-operatively, longer than reported previously (32).
There are a number of limitations in the present study. Only post-operative measurements were available for analysis and we cannot claim that the differences were solely a consequence of the surgical approach rather than being due to pre-operative functional differences. Furthermore, implant orientation/placement was not controlled for and it has been shown previously that this can affect hip function when performing ADLs (33). In an effort to control for this, patients were carefully selected to ensure that each surgical group had been operated on by a single experienced hip surgeon. Finally, the analysis of the CSA was only performed in a small subset of patients; increased numbers would provide a clearer understanding of the effect of surgical approach on muscle CSA. As this was only an exploratory part of the study these numbers were justifiable on pragmatic grounds and provide some future directions.