Hip & Pelvis

Cracking dysplastic hip arthroplasty

Metal-on-metal in the longer term

As the metal-on-metal debate continues to run, definition of what exactly is an acceptable outcome continues to ping-pong back and forth between supporters and detractors from the metal-on-metal (MOM) camp. This, the most recent study from Oxford (UK), is a further report from one of the most publicised series in the orthopaedic literature. With the longest follow-up published to date, the Oxford group have reviewed the outcomes following revisions after metal-on-metal hip resurfacings (MoM HR).² This series really does make grim reading. The reasons for failure following MoM arthroplasty are well documented, with certain designs reporting ten year revision rates of between 10%-13%. In total, 53 MoM HRs underwent revision with a mean patient age of 55 years,62% of whom were women. The most commonly revised implant was the Birmingham Hip Resurfacing (BHR) (55%) at a mean post-implantation interval of 1.6 years. The leading indications for revision were pseudotumour formation (30%) and femoral neck fracture (40%). Less commonly, cases of aseptic loosening, infection and recurrent dislocation were also revised. The majority of revisions were performed through a posterior approach, and both components were revised to a non-MoM bearing the majority (62%) of the time. All femoral stems were cemented and all acetabular components were uncemented. In the remaining 38%, only the femoral stem was revised for fracture, loosening or head collapse, for example. The femoral stem-only revisions were performed pre-2008 before complications with large head MoM THA bearings were known. A total of 24 patients (45%) undergoing MoM HR revision surgery sustained a complication and 20 patients (38%) underwent re-revision. Re-revisions were performed at a mean of 3 years following index revision and were most commonly for pseudotumours (40%), recurrent dislocation (20%) and deep infection (20%). The ten-year survival free from re-revision for all revised MoM HRs was just 63%. From the results of this study it was clear that those patients who had revisions for pseudotumours had inferior patient-reported outcomes compared with other revision indications, and the authors freely accept that the results of the femoral-only revisions were adversely affected by the use of large-diameter MoM bearings which have the worst implant survivorship. Like all retrospective cohort studies, the data should be taken with a small pinch of salt. Small numbers in subgroup analyses, surgical learning curves and not inconsiderable selection biases all have their roles to play here. However, this is the first long-term follow-up study of its kind, which gives it a certain significance and challenges some of the early studies reporting on the short-term results and patient-reported outcomes after MoM HR revision which suggested outcomes were comparable to conventional THA.

Hip injection – just what is the risk?

Any casual conversation in the theatre coffee room will clearly establish that different periods of time between an intra-articular steroid injection and a total hip arthroplasty (THA) are allowed by different clinicians. Intervals ranging from six weeks to 12 months could all be described as routine practice. Hip injections with local anaesthetic and steroid or viscosupplements can be both useful therapeutically to provide pain relief in hip arthritis, but also diagnostically if the patient has additional lower back pain as well as hip pain. However, there is always the concern that we may be increasing patients’ risk for a periprosthetic joint infection (PJI) when performed pre-operatively. It is therefore with great interest that we read this article at 360 – one of the few that cross the editorial desks with the capacity to change practice instantly. The authors in Charlottesville, Virginia (USA) set out to establish what a safe interval was following injection into the hip and THA without increasing the risk of infection.³ The difficulty of course, given the low event rates here, is the massive numbers needed to inform the study. The authors reviewed a total of 34 597 records of patients who underwent THA. These patients had been identified from an insurance-based database in the USA. These patients were then divided into three groups; THA within three months following an ipsilateral hip injection (829 patients), THA between three and six months after ipsilateral hip injection (1379 patients) and THA between six and 12 months (1160 patients) after ipsilateral hip injection. In addition, there was a control group of patients (31 229 patients) who had a THA but had never had an ipsilateral hip injection. There were no significant differences between the cohorts in all but one of the patients’ demographics, with a statistically significant higher percentage of female patients in the 0-3 month group compared to the control group. The incidence of infection after THA at three months and six months was significantly higher in patients who underwent hip injection within three months before THA compared with the controls. There was no significant difference in infection rates in patients who underwent THA between three and six months or six and 12 months after an ipsilateral hip injection compared with the controls. The literature to date has produced some conflicting evidence. There are a number of studies demonstrating no association between pre-operative hip injection and PJI after THA, but also similar numbers demonstrating higher rates of PJI. The biggest problem with all of these studies is the low numbers of patients involved, resulting in them being inadequately powered. Accepting the limitations of the study design, the quality of data extracted from the database and coding inaccuracies, this is the first study of this size that has given some clear guidance on when it is safe to proceed with a THA after an ipsilateral hip injection. Bottom line: wait at least three months.

Optimal evaluation of pincer type impingement

It is widely recognised that pelvic motion is a dynamic thing, and that pelvic tilt can affect positioning of the acetabulum and hence may impact on impingement. Authors from Pasadena, California (USA) undertook plain film evaluation of the pelvis in both standing and supine positions.⁴ The authors undertook radiographs on all new patients being evaluated for hip pain under 60 years of age. They then formally calculated the usual measures for femoroacetabular impingement, specifically intrapelvic distances (sacrococcygeal to symphysis [SC-S] and coccyx tip to symphysis [T-S]), crossover sign, LCE angle, inclination, and ischial spine sign. There were 46 paired radiographs suitable for inclusion in the study, and radiographs were evaluated by two independent observers. Pelvic tilt reduced from sitting to standing, with reduced T-S and SC-S distances. When evaluating both views there were no real differences in CE angle; however the ischial spine and crossover signs were seen significantly less frequently in the standing images. While films in the standing position are potentially more reflective of function, another consideration is that supine images are more reliably obtained by radiographers. There is certainly however food for thought here, as it does appear that at least a proportion of measures such as the crossover sign are in fact artefactual.

Can pre-operative hygiene reduce infection?

In the early days of arthroplasty, surgical patients were supervised in taking a pre-operative bath by the ward sister to ensure that they were appropriately clean for theatre. In these days of high turnover, same-day admissions and accelerated recovery pathways, there has been some loss of focus in many units on the basics such as perioperative hygiene. In patients with MRSA, or a high MRSA risk, there is a focus on pre-operative skin preparation, and surgeons in Brooklyn, New York (USA) have asked the question: would a preadmission chlorhexidine skin preparation reduce peri-operative infection? Their paper describes the outcomes of over 3500 patients, 998 who used chlorhexidine cloths pre-operatively and 2846 who did not.⁵ Subsequently a direct notes review was undertaken to establish which patients then went on to develop post-operative infections. There was a significantly higher infection rate in the ‘control’ group (1.6% vs 0.6%), however when the study team then went on to stratify the patients based on NHSN risk of infection characteristics, there were no differences between groups. This is a promising intervention – although the stratified analysis did not find any differences, it is important to remember the event rate in the intervention group is only 0.6% (equating to six patients). When subcategorising this by three risk groups, it becomes unsurprising that there were no tangible differences in event rates between the cohorts.

Femoral neck stress fracture

Ameliorating the systemic response to surgery?

Rationing based on national registry data?

Assessing outcomes in total hip arthroplasty

Perhaps brought into sharper focus by the metal-on-metal difficulties - although this paper applies to all branches of arthroplasty - researchers in Oxford (UK) have undertaken a systematic review with the express intention of establishing what constitutes the surrogate markers of long-term outcomes in hip arthroplasty.¹¹ With improvements in longevity and rising numbers of patients requiring intervention, failure itself may not be sensitive enough during the introduction of novel technologies to allow for the additional safety and monitoring that seems appropriate in the light of some high-profile failures. So, what is the gold standard of surrogate outcome measures in 2016 for monitoring changes to existing technologies or introduction of new arthroplasty technologies? The authors undertook an extensive review of the current literature and were able to identify 1082 studies, of which 115 were reviewed in full as fulfilling the inclusion criteria. The authors report on the findings of 17 papers, describing three approaches to surrogate outcome measures, and were able to conclude that there was enough evidence to describe both radiostereometric analysis (RSA) and Einzel-Bild-Röntgen-analysis (EBRA), both of which are able to measure both migration and wear as ‘validated’. The authors identified five RSA studies (one systematic review and four case series) and four EBRA studies (one RCT and three case series) supporting their use as surrogate outcome measures. However, the use of patient-reported outcome measures was not felt by the review authors to be suitable, as although potentially promising, they were not validated against longer-term outcome measures.