The mobilization process after bipolar hemi-arthroplasty can be improved, if a minimal invasive direct anterior approach is used as compared to a conventional approach.
There exist several studies comparing elective minimal invasive (MIS) hip arthroplasty with conventional approaches. The results of these studies regarding the clinical outcome, operating time and blood loss is often unequivocal. The outcome measurement systems used in these studies (e.g. Harris hip score [41
]) were all developed for a much younger, mentally fit and active patient population. Also these scoring systems were mostly constructed for a long term monitoring of the results of hip surgery.
In a recent systemic literature review, 14 commonly used outcome scales were found to be used for patients with proximal femoral fractures [18
]. None of these test methods were validated for use in this patient group. The author failed to identify a validated test method for this patient population. It remains unclear if the test methods that are used to detect the differences between minimal invasive and conventional hip arthroplasty (often HHS or WOMAC) might lack the discriminatory power in a short term study like ours; it is certain that these test methods are of limited use in a geriatric population. In a recent retrospective study, the period to successful mobilisation was measured in patients treated with a hemi-arthroplasty either with a conventional or a MIS approach [42
]. Apart from the fact that this was a retrospective study, the time point of successful mobilization was defined as the date at which the patient was able to stand coordinated with both legs and under use of available walking aids lift the non-operated leg from the floor. This item appears methodically difficult, especially in a retrospective study.
Although not validated for patients with a fractured proximal femur, the Barthel index has been used in trials before [21
For this study we decided to use an abbreviated version of the Barthel index. This method that has been validated in the past for geriatric neurological rehabilitation patients [27
]. We decided to use such an abbreviated test method with a focus on lower extremity function measured by activities of daily ling (ADL) in a geriatric population to increase the discriminatory power. This method is certainly not hip specific and we did not have the possibility to validate it previous to our trial.
Our results show an equal result at day 1 after surgery with the lowest possible 4-item Barthel index of median 0. This is not surprising as these frail patients often only get mobilized briefly on that day with the help of two physiotherapists. On day 5 and the following measurements, the DAA group showed superior results regarding the mobilization process. In the only published randomized trial comparing minimally invasive versus conventional hemi-arthroplasty for femoral neck fractures, full weight bearing was achieved faster in the minimal invasive group [43
]. Unfortunately this study neither specified the implants used nor how many were implanted cemented or uncemented.
Another randomized trial compared the mobility 48 hours after surgery and found no difference for the activities transfer from supine to sit, transfer from sitting to standing, mobilizing, ascending and descending stairs and weight-bearing for patients with elective THA [44
]. A recent trial comparing minimal invasive and conventional approaches for THA distinguished between anterolateral and posterolateral approaches [45
]. In this study, the superior results of the minimal invasive group regarding the HHS at 6
weeks were mainly found in the posterolateral access group. Other randomized trials failed to detect a significant difference in clinical outcome [46
]. Two meta-analyses failed to detect a significant difference between minimal invasive and conventional elective THA regarding the HHS [16
Operative time was about nine minutes longer for the minimal invasive group in our study, the difference between groups was not significant. Similar results have been reported before [43
], although possibly the operating time is shorter in minimal invasive procedures using a posterior approach [16
]. Unsurprisingly, the skin incision length in our study was shorter for the minimal invasive DAA group. We cannot fully exclude a bias as the length was only measured once.
A number of studies have measured intraoperative or postoperative blood loss [43
], the results are mixed. Even three meta-analyses came to divergent findings on this topic, showing either a highly significant advantage for a minimal invasive procedure [15
] or no significant difference [48
]. As the intraoperative blood loss is very difficult to measure exactly, we decided to determine the postoperative haemoglobin level, which showed no difference between the study arms. Also the measurement of administered packed red cell units was equivalent between the groups.
We found no difference in the direct postoperative pain using the VAS. Interestingly there was a detectable difference with less pain in the DAA group from day 5 onwards, which was still measurable at the end of the study on day 40. As the difference was small, it remains unclear if it is clinically relevant. A better pain control in a minimal invasive group also has been found in elective THA [17
The radiological analysis revealed no statistically significant differences between the two groups. Nevertheless no direct conclusions should be drawn from this fact for two reasons: First the study size was determined for the comparison of the 4-item Barthel index and it is possible that this prevented a difference from getting evident. Secondly the radiographic evaluation of plain radiographs for the measurement of hip arthroplasty is not very accurate [51
]. As this is mainly caused by systematic errors, both groups should be influenced about equally.
According to the German “BQS national quality report in orthopaedics and traumatology”, a 30
day mortality rate of 5.9% has been reported in 45,051 patients treated with an endoprosthetic device for a fractured neck of femur in 2007 [53
]. In our study with no mortality, frail and bedridden patients were excluded as the primary outcome item measured mobility. In a study with a similar patient collective, two of 69 patients died (2.8%) due to pulmonary embolism [54
]. Regarding intra- and postoperative complications we found no evidence that a minimal invasive approach results in a higher complication rate. All surgeons involved in the treatment in our study were on a senior level with a long experience in hip fracture treatment; also they underwent cadaveric training by the inventors of this method previous to the trial. The results regarding intra- and postoperative complications might therefore not be directly applicable to various hospital settings. Complication rates of this approach have only been described for elective hip surgery [9
], it is unlikely that these are lower in patients with a fractured neck of femur.
The strength of the present study is that it is a prospectively randomized trial with a single type of implant and a homogenous surgical method. Apart from the type of approach, the treatment scheme was identical and the population of the groups was comparable.
The present study is not able to address the question if an improved mobilisation results in fewer complications or a lower mortality, this is certainly a weakness. Much larger groups and a longer follow up period would be required to answer for this. There was no external monitoring of this study; as a result we only provided level 2 evidence with this trial.