Aseptic Loosening

Murray J. Penner

Sulaiman A. Almousa

Lee Kolla

INTRODUCTION

The leading cause of failure and subsequent revision surgery following total ankle arthroplasty (TAA) is aseptic loosening.¹ However, exact rates of aseptic loosening are difficult to define, since definitions and methods of diagnosing loosening vary considerably in the literature.

In general terms, aseptic (i.e., not caused by infection) loosening refers to the failure of fixation at the bone-implant interface, with resultant micro- or macromotion of the implant relative to the adjacent bone. The challenges in diagnosis arise from the difficulty in detecting the presence of such motion, particularly when it is in the submillimeter range. Thus, surrogate measures, such as radiolucent lines adjacent to implants or increased uptake on bone scan, are typically used to determine whether an implant is loose. However, the accuracy and interobserver agreement of these measures are unknown.

Loosening may occur early, through failure of initial ingrowth of bone into the prosthesis or poor cementing technique. Alternately, loosening of a previously solidly fixed implant may occur months or years after implantation, potentially because of mechanical overload, physiologic bone resorption, or a combination of both at the bone-implant interface. This leads to the varied clinical presentations associated with loosening, which may range from no pain to persistent ankle pain beginning immediately after TAA to late-onset pain beginning many months or years after a previously nonsymptomatic TAA.

Further complicating matters is the recognition that implants may be partially loose. Just as recent computed tomography (CT) scan studies have shown that joint arthrodeses often have bone-to-bone healing across only a portion of the joint surface,² it is also apparent that bone ingrowth may occur only over a portion of the bone-implant interface. Depending on the location and amount of ingrowth present, a large enough portion of the implant may not be fixed to bone, thereby creating a cantilever effect, much like a diving board, where one side of the implant is stable, but the other side experiences micromotion.

It is these complexities that make a thorough understanding of aseptic loosening a challenge. The goal of this chapter is to define the etiology, epidemiology, and classification and diagnostic approach to aseptic loosening in TAA, within the limits of this challenging context.

EPIDEMIOLOGY

Aseptic loosening, with or without implant subsidence, is the leading cause of TAA failure.¹^,³ It is notable that some authors categorize aseptic loosening and subsidence separately, while others do not. This has the potential to create confusion, since criteria to separate these two categories are not universal. In general terms, subsidence refers to macroscopic motion of the implant relative to bone, a condition that inherently indicates that the implant is loose, while aseptic loosening implies nonmacroscopic loosening. Hence, both of these terms represent loosening, and for the purposes of this chapter, both will be considered.

Glazebrook et al.¹ performed a systematic review of articles reporting on TAA complications and failures. They included all cohorts with at least 25 patients and minimum 2 years of followup. They reported a mean failure rate of 12.4% (range, 1.3% to 32.3%) at 64 months for the 2,386 ankles reviewed. Aseptic loosening and subsidence (i.e., macroscopic loosening) were the most common complications, with a combined rate of 19.4% (10.7% and 8.7%, respectively). Aseptic loosening resulted in failure of the TAA 70.2% of the time that it occurred. On the basis of this rate of failure, they classified aseptic loosening as a high-grade complication, along with deep infection and implant failure.

Haddad et al.⁴ performed a meta-analysis pooling TAA outcomes in 10 intermediate- to long-term studies evaluating a total of 852 TAAs. They reported a revision rate of 7% (95% confidence interval, 3.5% to 10.9%) at a mean of 4.7 years postsurgery, with the primary reason for the revisions being loosening and/or subsidence. However, the 5-year survival rate was only 78%.

In a recent study with the longest-term follow-up for any contemporary TAA prosthesis, Brunner et al.⁵ found aseptic loosening and subsidence requiring revision in 20 (32%) of 62 Scandinavian total ankle replacement (STAR) cases available for follow-up at a minimum of 10.8 years. Younger age at the time of TAA was associated with an increased risk of loosening.

In one of the earliest studies on a contemporary TAA in North America, Pyevich et al.⁶ found that 21 of 85 Agility TAAs had migrated (i.e., were macroscopically loose), although only two underwent revision. Radiolucent lines of 2 mm or less at the bone-implant interface were found circumferentially around the tibial component in 26% of cases. The authors noted that
these lines were always present within 2 years of implantation, suggestive of failure of initial bone ingrowth.

In summary, aseptic loosening is the most common major complication following TAA, commonly leading to revision surgery. Loosening may occur early, possibly related to failure of initial fixation, or later, with increasing frequency over time. Rates of loosening vary widely between studies, from approximately 5% to 30%, and clear correlation with implant type or other factors is not fully defined. Nonetheless, younger age at the time of TAA and increasing length of time after TAA do appear to be strongly associated with increased risk of loosening.

ETIOLOGY

Little has been written about the etiology of aseptic loosening in TAA specifically. However, aseptic loosening has long been recognized as a major complication of total hip arthroplasty (THA) and total knee arthroplasty (TKA). As a result, most of what is understood about aseptic loosening is taken from THA and TKA literature and extrapolated to TAA. Although the validity of such a wholesale assumption is debatable, many, if not all, of the same contributory principles present in THA and TKA are also present in TAA in some manner. As a result, this section relies primarily on data from THA and TKA experience, while incorporating TAA-specific data where possible.

Shortly after the introduction of prosthetic joint replacement, periprosthetic bone loss (osteolysis) and eventual component loosening were recognized as a main mode of implant failure. Several theories have been postulated to explain this phenomenon. Initially it was attributed to chronic osteitis secondary to mild sepsis⁷ or to hypersensitivity reaction to cement also called “cement disease.”⁸^,⁹ It is now believed that biologic reaction to particulate wear debris plays a central role in the pathogenesis of osteolysis and aseptic loosening of various prosthetic joints. In general, aseptic loosening could result from a harmful combination of mechanical and biologic factors that jeopardize the formation or the survival of bonding between the implant and the host bone.¹⁰ These factors could be divided into six broad categories: (1) biologic response to wear debris, (2) intra-articular fluid pressure, (3) implant design, (4) patient-specific characteristics, (5) dormant unrecognized infection, and (6) genetics.

BIOLOGIC RESPONSE TO WEAR DEBRIS

The role of particulate debris in periprosthetic bone loss has been studied extensively since its initially recognition by Willert and Semlitsch¹¹ in 1977. Biologic response to wear particles is now recognized as the leading cause of periprosthetic osteolysis and aseptic loosening of prosthetic hip and knee implants. Polyethylene liners, metal components, and cement all are subjected to wear and produce wear particles. Of these, polyethylene particles are the most important in the pathogenesis of osteolysis. Particle type, size, and number affect the host biologic response.¹²^,¹³ Green et al.¹² describe a “critical size” range (0.3 to 10.0 µm) that the wear particles must fall within in order to trigger a macrophage-based inflammatory response. Critical-size wear particles are phagocytized by macrophages, triggering a cascade of intracellular reactions leading to the production of inflammatory mediators, including tumor necrosis factor (TNF)-α, interleukin (IL)-1, IL-6, and macrophage colony-stimulating factor (M-CSF). TNF-α induces fibroblast proliferation, tissue fibrosis, and activation of osteoclasts. This leads to extensive periprosthetic bone resorption.¹⁴^,¹⁵ As well, wear debris has a direct inhibitory effect on osteoblast bone formation.¹⁶

Schmalzried et al.¹⁷ introduced the concept of effective joint space. The effective joint space is the space surrounding the prosthetic joint and encompassing all of the implant-bone surfaces through which synovial fluid can flow and disperse wear particles. This concept explains how wear particles reach areas far from the articular surface.

INTRA-ARTICULAR FLUID PRESSURE

Inflammation triggered by wear debris particles and/or intraarticular exposure of bone which is normally sealed from the joint results in overproduction of synovial fluid and a potential increase in intra-articular fluid pressure; this, in turn, may result in abnormal bone perfusion and ischemia leading to necrosis, osteocyte death, and osteolysis. This effect was demonstrated in animal experimental studies.¹⁸^,¹⁹ Robertsson et al.²⁰ documented a higher intra-articular pressure in 18 hips diagnosed with aseptic loosening in comparison to stable hips.

IMPLANT DESIGN

Immediate implant stability is critical in achieving strong bony ingrowth at the bone-implant interface²¹ and failure of ingrowth may lead to early aseptic loosening of the prosthesis. Younger et al.²² have demonstrated significantly greater micromotion immediately after implant insertion for Agility TAA implants compared to STAR implants, and have correlated this with a significantly higher rate of revision due to aseptic loosening in Agility TAAs compared to STARs. This finding suggests that the early circumferential radiolucent lines identified by Pyevich et al.⁶ in 26% of Agility TAA cases may be due to failure of initial bone ingrowth secondary to insufficient initial implant stability.

Achieving solid initial implant fixation may be affected by several implant design factors. The addition of keels or stems to the implant, for example, may provide a larger fixation surface, increasing the implant initial stability, reducing micromotion and mechanical stresses at the bone-implant interface, thereby potentially increasing the chance of successful bonding.²³^,²⁴ Ries et al.²⁵ retrospectively compared standard and short-keeled TKA and showed an increased risk of aseptic loosening in the short-keeled TKAs.

For stemmed implants, increased stem flexibility may result in decreased bony ingrowth, increased fibrous ingrowth, and increased risk of implant loosening.²⁶ On the other hand, increased stem stiffness may result in more stress shielding and periprosthetic bone loss.²⁷^,²⁸

Adding a porous or hydroxyapatite-coated surface to the implant may help seal the implant-bone interface. This seal prevents the wear particles from reaching the effective joint space surrounding the implant and reduces wear debristriggered periprosthetic osteolysis.²⁹,³⁰,³¹ and ³² The presence of unused screw holes may provide a portal for wear particles to flow into the surrounding bone and may increase the risk of osteolysis³³ and eventual subsequent loosening.

Only gold members can continue reading. Log In or Register to continue