Introduction

Hemiarthroplasty of the hip is the treatment of choice for displaced intracapsular fractures of the neck of femur in the elderly. A number of prostheses have been used since the inception of this treatment, a wide variety of which are currently available in both cemented and uncemented versions. Cementing stems into the femur has a number of effects on the cardiovascular system, recently termed bone cement implantation syndrome [1], which are potentially deleterious to elderly patients. The use of uncemented hemiarthroplasties to avoid the physiological effects of cementation has obvious appeal. This has been mitigated by the perceived increase in other complications, such as periprosthetic fracture [1–5]. At our institution, the authors have been using the Taperloc hydroxyapatite (HA)-coated uncemented prosthesis (Biomet, US), for the treatment of all intracapsular neck of femur fractures requiring femoral head replacement since 2007. A retrospective review of this stem to evaluate the incidence of intraoperative fracture and whether surgical experience affected this was undertaken. The postoperative complications and the 30-day mortality rates were also examined.

Materials and methods

All trauma operation details are recorded prospectively onto our trauma database by the operating surgeon. The database subjects were interrogated for all hemiarthroplasties performed using the Taperloc prosthesis from September 2007 to July 2009. The prosthesis comes in 10 sizes, with standard and lateralized types available, and has a proximally HA-coated tapered wedge design. In all cases, the procedure was performed through a lateral approach with the patient in the lateral decubitus position. After broaching the femoral canal, rasps of ascending sizes are used. The rasp that obtains good rotational stability in the femoral canal, (rather than aiming for a complete press-fit), determines the eventual stem size used. Following stem insertion, a modular unipolar hemiarthroplasty head is attached before hip reduction. Postoperatively, all patients were mobilized fully weight bearing and a postoperative radiograph was taken. All postoperative imagery was reviewed using the PACS system.

Results

A total of 172 procedures were performed. The study was carried out at a mean of 1.1 year after surgery, (range 0.4–2.1 years). No patients were excluded from this series, and this prosthesis was solely used during the period studied. The age range was 60–103 with a mean of 84 years, and the mean time from admission to surgery was 3 days. The operations were carried out by surgeons of varying experience. Specialist registrars performed 71%, with half of these supervised by a consultant. 24% were performed by consultants, and 5% by senior house officers. All patients had a postoperative radiograph within a week of their operation. No periprosthetic fractures were sustained as a result of stem insertion, and none of the patients was later readmitted to our institution for a periprosthetic fracture. There were no intraoperative deaths.

There was one intraoperative greater trochanter fracture sustained during relocation of the hip. This was wired back on at the same procedure. Four patients had complications after surgery. Two patients had a revision to total hip replacement, one following a dislocation 4 months after insertion, (Fig. 1), and one following acetabular wear and superior migration 9 months after insertion, (Fig. 2). In both cases, the original stem was retained. Two other prostheses were converted to a Girdlestone, one–two months postoperatively for infection, and one following recurrent dislocation due to poor quality soft tissues 2 months after insertion, (Fig. 3). Over the period studied, the 30-day mortality for these patients was 4%.

Fig. 1
figure 1

Radiographs showing the dislocated prosthesis 4 months after surgery and the postoperative radiograph following conversion to total hip replacement using the original stem

Full size image
Fig. 2
figure 2

Radiographs showing superior acetabular wear and prosthesis migration at 9 months, alongside the postoperative radiograph following conversion to total hip replacement, with the original stem retained

Full size image
Fig. 3
figure 3

Radiographs showing the dislocated prosthesis and eventual conversion to a Girdlestone due to poor quality soft tissues

Full size image

Discussion

The cementing of femoral stems can lead to a gradation of effects: Hypoxia, hypotension, increased pulmonary vascular resistance and pulmonary shunt, cardiac arrhythmias and ultimately cardiac arrest have been recognized [1, 6, 7]. The exact aetiology of this effect has been the subject of some debate, but the likely explanation is the formation of emboli due to increases in intramedullary pressure seen at cementation and prosthesis insertion [8]. The presence of these showers of emboli has been shown to correlate with alterations in cardiovascular physiology [9, 10].

Whatever the underlying cause, these cardiovascular events can be fatal in the perioperative period. Although the reported incidence in large cohorts is low for intraoperative deaths [11], the physiological insult at the time of surgery may increase the perioperative mortality risk as well [12], although there is conflicting evidence concerning this [4]. Modern cementing techniques have been developed to reduce the potential problems, and may reduce the incidence of clinically detectable events [13].

However, the cardiovascular changes are still observed even if these techniques are used [7], and patients undergoing emergency treatment for a fractured neck of femur are very different cohort from those undergoing elective THR. They are often elderly with multiple comorbidities and a limited physiological reserve, further inhibited by multiple medications, with which to respond to cardiovascular derangement. Although the reported incidence of perioperative events relating to cemented hemiarthroplasty is low, the voluntary nature of the reporting system in the UK may well underestimate the true numbers [14]. The use of uncemented stems, especially in high risk patients is an alternative option. However, there are concerns with the increased risk of perioperative fracture with many designs [2–4, 15], and intraoperative fracture rates of up to 14% have been reported for prostheses in widespread use, independent of the experience of the surgeon, [5].

The most recent Cochrane review identified six trials comparing cemented and uncemented prostheses [16]. Those authors concluded that the available evidence suggested that cemented prostheses gave improved pain and mobility outcomes. However, they acknowledge that the studies contain significant heterogeneity of prostheses used and problems with design. In addition, they have excluded trials using modern HA coated stems from this part of the analysis. There have been two recent prospective randomized trials. Figved et al. [17] compared hemiarthroplasties using two stems with excellent track records in total hip arthroplasty, and did not demonstrate a significant difference in functional outcome or complication rates between the cemented or uncemented groups. A comparison of cemented Thompson hemiarthroplasty with an uncemented Austin-Moore prosthesis [18] demonstrated less pain and better function in the cemented group, without a significant difference in complication rates. However, despite the statistically comparable mortality, this article did record one intraoperative cardiac arrest in the cemented cohort, and excluded 39 patients from the trial who were considered unfit for a cemented procedure. The clinical impact of bone cement implantation syndrome is difficult to quantify, but many units still use uncemented hemiarthroplasties in certain patient groups considered to be high risk, and currently only 63% of prostheses used in the UK are cemented in place [19].

The Taperloc stem used in our institution has excellent long-term results in total hip arthroplasty [20, 21]. The authors feel the greater preferred choice of stem options and more forgiving tapered designs, coupled with an operative sizing technique designed to gain stability without going for absolute press-fit has led to the absence of periprosthetic fractures as a result of stem insertion in our series. This remains the case even in situations where the degree of osteoporosis of the bone requires a larger prosthesis, (Fig. 4). The 4% 30-day mortality of the patients in this cohort is well below the average of 7.8% reported throughout the UK [19]. This latter statistic obviously involves many more factors than the choice of stem, but it is clear that the exclusive use of the Taperloc stem does not adversely affect this important outcome measure. Our series demonstrates that uncemented hemiarthroplasty with this prosthesis can be safely and reliably performed by surgeons of varying experience in all patients requiring replacement of the femoral head following a fractured neck of femur without significant risk of periprosthetic fracture or increase in peri-operative mortality. The authors continue to use this prosthesis for all patients at our institution and would recommend it to other surgeons as a better option to older monoblock versions in situations where an uncemented prosthesis is preferred. There is of course a cost implication to using modern technology, but this should be weighed against the potential costs of complications following surgery.

Fig. 4
figure 4

Postoperative radiograph. The osteoporotic nature of the bone required the use of a size 20 prosthesis, the largest size available, with satisfactory stability at surgery

Full size image