Hip & Pelvis

Body mass index and mortality after total hip and knee arthroplasty

Ceramic-on-ceramic or ceramic-on-polyethylene: 15-year follow-up

At present, it remains unclear whether younger patients are better off with a ceramic-on-polyethylene (CoP) or ceramic-on-ceramic (CoC) bearing hip arthroplasty. This Canadian randomized trial from Toronto (Canada) has previously been published up to ten-year follow-up, with their previous report showing no difference in either revision rates or symptomatic/functional outcomes. The same group now publish their minimum 15-year results.² The authors undertook recruitment to this randomized trial between October 1997 and 1999, where 58 primary hip arthroplasties were undertaken for osteoarthritis in 57 active patients aged under 60 years. All patients were randomized to receiving either CoC or CoP (29 vs 28 patients; one had bilateral surgery). This series reports on the longer-term follow-up for this group of young and active patients. The overall 15-year revision rate of 16% is higher than might be expected in more recent series, but the authors make a point that both polyethylene and ceramic manufacture have evolved since this particular series was undertaken (higher cross-linkage and delta ceramic, respectively). Allowing for the three patient deaths (5%) and seven patients (12%) who were lost to follow-up, of the remaining cohort, there was no statistical difference between revision rates with CoC versus CoP (four and five, respectively). Harris Hip Score and St Michael’s Hip Score were comparable between the two bearing surface groups. The numbers are small in this study. However, this does represent one of the few randomized trials undertaken with differing bearing surfaces. On the basis of these data, at a minimum 15-year follow-up, there appears to be no significant clinical difference between the two bearing services. The authors did find, however, that differences were demonstrated in observed wear rates between the two bearing surfaces. All patients in the CoP cohort had a degree of detectable wear, compared with just 12 of those with CoC. Not unreasonably, the authors observe that this may be a portent of potentially higher failure rates at a longer follow-up with hard-on-soft versus hard-on-hard bearings, although they also postulate in their discussion that this might, to an extent, be reduced with the advent delta ceramic heads and ultra-highly cross-linked polyethylene in the longer term. The main value of the present study, however, is that, even whilst allowing this degree of license for more recent changes in both the polyethylene and ceramic components, these data suggest that both bearing couplings can justifiably be used in younger, more active patients, with the expectation of equivalent survivorship between the two cohorts, at least up to 15 years.

A UK FASHIoN? Hip arthroscopy in a randomized controlled trial

This is a well-designed, high-impact, assessor-blinded study originating in Coventry (UK) comparing the 12-month results of personalized physiotherapy versus hip arthroscopy for femoroacetabular impingement (FAI).³ Following an initial identification of patients thought potentially to be likely to benefit from hip arthroscopy for FAI (either cam or pincer), those with established OA Tönnis (Grade I or above) were excluded, as were those with previous hip pathology (e.g. Perthes’ disease, slipped upper femoral epiphysis, avascular necrosis, trauma, or previous surgery). The authors were able to recruit 348 patients from across the 23 participating hospitals, and randomized them to receive either arthroscopy (n = 171) or physiotherapy (n = 178). The 12-month follow-up rate was 92%. The primary outcome measure was hip-related quality of life, quantified with the international Hip Outcome Tool (iHOT-33). Both physiotherapy and arthroscopy groups showed improvements 12 months after randomization. A number of different sub-analyses were undertaken; in all of these (most notably the intention-to-treat analysis), hip arthroscopy showed statistically significantly greater improvements in iHOT-33 than physiotherapy, even when allowing for the fact that 73% of the hip arthroscopy patients reported some form of adverse event. These adverse events were most commonly muscle soreness, but also various postoperative complications, including infection, scrotal haematoma, and so on. It is an important detail that the mean difference in iHOT-33 scores was interpreted adjusted for impingement type, sex, baseline iHOT-33 score, and centre, which is slightly unusual, as, due to their randomized nature, trials such as this do not usually need an adjusted analysis. The research group conclude that offering hip arthroscopy patients with femoroacetabular impingement syndrome leads to superior patient-assessed function 12 months following randomization, compared with the (perceived) best nonoperative treatment. They acknowledge, however, that this does not equate to better cost-effectiveness of hip arthroscopy over care within the first 12 months. This an important addition to the ever-expanding literature on this topic, and represents significant support for the use of arthroscopy in the treatment of this patient cohort.

Fixed spinopelvic alignment and hip dislocation

The cause of hip instability following total hip arthroplasty (THA) is often multifactorial, being associated with surgical factors such as operative approach, choice of implant, implant positioning, and soft-tissue repair. Of these factors, it is often the implant positioning that surgeons spend considerable time and effort trying to get right. Previous literature has referred to the ‘safe zone’ of acetabular placement. However, siting the acetabular component in the ‘safe zone’ does not guarantee stability. The reason for this is that the orientation of the acetabular component is not static and changes according to the position of the pelvis. Pelvic orientation changes according to whether the patient is standing, sitting, or lying flat during the gait cycle. This becomes more complicated when the patient has had previous spinal surgery, such as a lumbar fusion, or suffers from a degenerate lumbar spine with little movement. Patients with stiff lumbar spines have little ability to change their pelvic tilt from standing to the sitting position. To be able to sit, patients must deep flex their hips and, in patients who have had a THA, this could result in rim impingement and a posterior dislocation. At the recent American Academy of Orthopaedic Surgeons meeting, and also at recent national meetings in the United Kingdom, there has been considerable discussion on how best to manage these patients and where to place the acetabular component. This study published from the Hospital for Specialist Surgery in New York, New York (USA) attempted to compare patients who have undergone a THA and have dislocated with those who have not, in terms of the lumbar-pelvic-femoral alignment in the sitting position in both groups.⁴ A total of 1000 patients with a mean age of 61 years were included in this study, who underwent a THA using 28 mm, 32 mm, or 36 mm femoral heads for osteoarthritis, all undertaken via the posterior approach. Importantly, the surgeons either used an imageless optical navigation system (AchieveCAS; Smith & Nephew, Memphis, Tennessee) or CT-based navigation (Mako Surgical, Tampa, Florida) to aid the implant positioning. The acetabular component was placed in 40° inclination and 20° anteversion. Patients were followed up for a minimum of one year. An imaging system was used so that patients could undergo standing and sitting radiographs from the thoracolumbar junction to the ankles using a low-dose radiation imaging system. This enabled the authors to visualize the entire lumbar-pelvic-femoral complex of patients in sitting and standing positions. From the reported series, a total of 12 patients (1%) experienced a posterior hip dislocation in the first year after THA. Of these 12 patients, 11 had multilevel lumbar degenerative disc disease, with four of these patients having undergone a surgical spine fusion prior to THA. Only one of the patients who dislocated had a ‘normal’ spine, although the authors demonstrated that this patient had more spine flexion and change in pelvic tilt than the other patients who had dislocated. As would be expected, the dislocating patients had significantly less spinal flexion, lower pelvic tilt, and more hip flexion from standing to sitting positions compared with the patients with normal spines. These patients had a lower sitting acetabular functional inclination angle and a lower sitting acetabular functional anteversion angle compared with the non-dislocators. The authors did not consider the femoral version, which has also been shown to influence hip stability post-THA. The problem of a relatively fixed spinopelvic alignment has gained considerable exposure in the more recent orthopaedic literature. This condition can be secondary not only to spinal fusion, but also to lumbar multilevel degenerative disc disease. This study demonstrated that patients with a fixed spinopelvic alignment from standing to sitting are more at risk of THA dislocation. It is important for hip surgeons to fully appreciate that conventional acetabular orientation does not then necessarily exclude the risk of dislocation. This paper is the first to formally describe this relationship between reduced spine flexion and increased hip flexion, and therefore higher risk of dislocation, post-THA. Rather than aiming for a conventional acetabular alignment of 40° inclination and 20° anteversion, there is a sound argument based on this paper that surgeons should opt for a functional acetabular orientation that will be different for different patients depending on the mobility of their lower spine. In those patients who are considered at risk, preoperative x-rays, including sitting and standing views, could therefore become the norm to fully understand what happens when patients stand and sit. Increasing the anteversion in patients with relatively stiff lumbar spines would require additional cup anteversion to reduce the risk of dislocation. However, the authors were unable to recommend an alternative ‘safe zone’ in patients with stiff lumbar spines, as each patient’s ‘stiffness’ was different to another patient.

Outcomes after metal-on-metal hip revision surgery

Metal-on-metal hip arthroplasties (MoMHRs), including stemmed hip arthroplasties and hip resurfacings, have in some instances experienced high rates of unexpected short-term complications requiring revision surgery. The reasons for revisions include dislocation, loosening, infection, fracture, and complications in relation to adverse reactions to metal debris (ARMD). In the case of ARMD, the complication rate of revision surgery has been reported as high (50%), likely due to the destructive nature of the lesions. This led to surgeons lowering their thresholds for revision surgery for this indication, with the aim of revision prior to too much soft-tissue damage. Revision surgery is a significant undertaking for both the patient and the surgeon, and robust data to support decision-making can be very helpful. The authors of this study reviewed a large cohort of patients obtained from the National Joint Registry (NJR) by researchers in Oxford (UK) who had undergone MoMHRs to compare the risk of complications and reoperation after revision surgery for ARMD and non-ARMD revisions.⁵ A total of 4908 patients underwent a primary MoMHR − 2913 MoM total hip arthoplasties (THAs) and 1995 hip resurfacings (HR) – and subsequently required revision surgery for any indication. To reduce the risk of bias, revision procedures were matched for multiple potential confounding factors using propensity scoring methods. The final matched cohort included 2576 MoMHR revision procedures, with 1288 in the ARMD group and 1288 in the non-ARMD group. The mean follow-up in both groups was three years. In terms of intraoperative complication at the time of revision, there was no difference between ARMD (2.4%) and non-ARMD (2.5%) revisions. The most common complication in both groups was calcar and greater trochanteric fractures. Excluding infection, there was no difference in mortality rates at five years between the two groups. All-cause re-revision rates were lower in the ARMD group (94.3% five-year survival rate) compared with the non-ARMD group (90.5% five-year survival rate). Re-revision indications included dislocation/subluxation, ARMD, infection, loosening with or without lysis, unexplained pain, and fracture. Revisions performed for infection and dislocation/subluxation had the lowest five-year implant survival rates compared with the ARMD group. This study suggests that those patients undergoing revision for ARMD had approximately half the risk of further revision and death compared with matched patients undergoing revision for other reasons. Importantly, they identified that infected cases were responsible for the higher mortality risk and infection, and dislocation/subluxation was responsible for the higher re-revision risk. This is not what previous studies have suggested, as revision for ARMD was thought to have a much higher risk of complications and further revision surgery. This study gives some encouragement to both patients and surgeons alike that revision surgery in the presence of ARMD is perhaps not quite as bad as was first thought. Clearly, careful planning is needed before any surgery, and the information from this study can help inform the discussion between surgeon and patient when considering revision surgery for MoMHR. Previously, surgeons have been encouraged to perform revision surgery at an early stage following the development of ARMD complicating MoMHR. This study would suggest that the threshold for revision should not be lowered any further, as there could be a risk of performing revision surgery too early, with the potential for the surgical risk outweighing the benefits. What is concerning is the high risk associated with revisions for infection and subluxation/dislocation, and further research is needed not only into the treatment of this condition but also, more importantly, into its prevention.

Reaction time and brake pedal depression following arthroscopic hip surgery: a prospective case-control study

“When am I safe to drive, doctor?” is a tricky question to answer. With little in the way of agreed national guidance from regulators, healthcare providers, or motor insurers, it is often left to the best guess of the clinician. Although there is a plethora of work now surrounding driving in plasters and following total joint arthroplasty, there is little that is known about many other relatively common procedures. A research team in Bethesda, Maryland (USA) used the now-familiar approach of a car simulator and measured braking response time following hip arthroscopy in an attempt to establish when exactly patients are safe to drive.⁶ The authors undertook a prospective cohort study of 59 patients undergoing arthroscopic hip surgery for femoroacetabular impingement (FAI) and compared these with 59 age- and gender-matched controls without FAI syndrome. Total brake reaction time (BRT) and brake pedal depression were measured preoperatively, and at two, four, and six weeks postoperatively. The group of patients with FAI had significantly prolonged BRT prior to surgery. In patients undergoing left hip surgery, there was no difference in any measures, whilst patients undergoing right hip surgery had significantly prolonged BRT at two weeks compared with preoperative baseline (688 vs 573 milliseconds). This paper, in spite of its limitations, provides information about brake reaction time prior to and following hip arthroscopy. The testing was done on a simulator and revealed a prolonged brake reaction time in patients undergoing arthroscopic surgery of the right hip until four weeks postoperatively. In general, it would therefore seem sensible to advise patients to avoid driving for a month postoperatively.

15-year prospective UEFA Elite Club Injury Study

Aspirin or rivaroxaban for venous thromboembolism prophylaxis after hip or knee arthroplasty