Results of Modern Total Ankle Arthroplasty

Samuel B. Adams Jr

Mark E. Easley

INTRODUCTION

Total ankle arthroplasty (TAA) is becoming a popular alternative to ankle arthrodesis for the treatment of end-stage ankle arthritis. Although almost abandoned because of high failure rates with historical prostheses, modern implants are demonstrating a dramatic improvement in clinical outcomes and survivorship. Additionally, pain subscores and functional outcomes are equal to and may exceed those of ankle arthrodesis.¹ Likewise, TAA has been shown to be cost-effective compared to ankle arthrodesis.²^,³

This chapter summarizes the results of modern TAA, with special attention to patient-specific factors and implant design. However, with more than 40 different implant designs and a myriad of outcome measures, direct comparison among reports of survivorship can be difficult. Additionally, confounding variables such as a prolonged learning curve for surgeons implanting TAA and changes in prosthetic designs and operative techniques make it difficult to sort out the reasons for success or failure. Likewise, the term “failure” is not used universally. In general, failure is defined as removal of one or both metal components with subsequent revision ankle arthroplasty or conversion to ankle arthrodesis. In some survivorship analyses, polyethylene meniscal bearing exchange and other reoperations are included as failures. However, reoperation in TAA does not imply failure. In fact, some repeated surgery for relieving impingement, improving alignment, bone grafting cysts, and/or exchanging the polyethylene component is anticipated to prolong implant survival.

META-ANALYSES AND SYSTEMATIC REVIEWS

The vast majority of TAA studies are case series reporting results of a single implant. However, there have been a limited number of meta-analyses performed to summarize the outcomes of TAA with multiple prostheses. Most studies examined in TAA meta-analyses fail to meet methodological standards, lack key data elements, and contain variability in operative procedures, evaluation tools, and reporting of outcomes. Therefore, generalizations about TAA outcomes derived from meta-analyses of the existing body of literature for TAA are difficult to make in a reliable manner.

Gougoulias et al.⁴ identified 13 eligible studies comprising 801 mobile-bearing and 304 fixed-bearing prostheses. Including only results of current TAA designs, the authors noted that functional outcome scores improved in all studies and there was no superiority of one prosthesis over another. An important point raised in this study was that residual pain was common after TAA, ranging from 2*7% to 60%. More importantly, chronic pain was present in patients with clinically and radiographically acceptable implants and sometimes prompted conversion to arthrodesis.⁵^,⁶

Stengel et al.⁷ identified 18 eligible studies of mobile-bearing TAAs with a mean follow-up of 44.2 months (range, 35.9 to 52.4 months). Using pooled data of outcome measures with 100-point scoring, they demonstrated a mean improvement of 45.2 points following TAA. A secondary surgery was performed in 12.5% of patients and arthrodesis was necessary in 6.3% of patients. They reported a 5-year survivorship of 90.6 (based on only six studies that actually reported survivorship data). The vast majority of reported results were for one particular design, and the authors did not identify differences in performance ratings when comparing that design against the other prostheses analyzed.

The systematic review by Haddad et al.⁸ of both TAA and ankle arthrodesis deemed 10 studies with 852 TAAs (both mobile-bearing and fixed-bearing designs) and 39 studies with 1,262 ankle arthrodeses eligible for systematic review. The mean American Orthopaedic Foot & Ankle Society (AOFAS) scores were 78.2 and 75.6 points for TAA and ankle arthrodesis, respectively, with mean subscores of pain, function, and alignment for TAA being 34.5, 37.4, and 9.4 points, respectively. For TAA studies with patient assessment that includes details of excellent, good, fair, and poor, mean results of meta-analysis were 38%, 30.5%, 5.5%, and 24%, respectively. These percentages were also calculated as a ratio of the number of patients with a particular outcome divided by the total number of patients reporting outcome: 48/92 (52.2%) excellent, 28/92 (30.4%) good, 4/92 (4.3%) fair, and 12/92 (13.0%) poor. In studies lacking such detail for TAA, good results were noted in 388/482 (80.5%) and poor results were observed in 94/482 (19.5%). With respect to the low patient satisfaction, their analysis included results of one study of a fixed-bearing two-component prosthesis⁹ that fall far below those of most studies of TAA,
including other studies of the same prosthesis.¹⁰^,¹¹ Likewise, their overall reported survivorship was lower than those reported in other meta-analyses. This is secondary to low implant survival reported in two studies.¹²^,¹³ One of these studies was by Anderson et al.¹² reporting a 5-year survivorship of 70% in a series of Scandinavian total ankle replacement (STAR) implants. The authors of that study acknowledged that their results and survivorship were negatively influenced by a so-called learning curve and an incomplete inventory of talar component sizes for their initial cases. Furthermore, the prosthesis they used did not consistently have the titanium spray, which is currently used on the mobile-bearing prosthesis that they were investigating. Likewise, the comparative study by Kofoed and Lundberg-Jensen¹³ included now abandoned early cemented and uncemented versions of the same mobile-bearing prosthesis.

Zhao et al.¹⁴ performed a systematic review on the available literature of STAR prostheses. They identified 16 studies representing 2,088 implants with a mean follow-up of 52 months. The mean AOFAS and Kofoed scores were 77.8 and 76.4 points, respectively. The pooled 5- and 10-year survival rates were 85.9% and 71.1%, respectively. This systematic review also included the study by Anderson et al.,¹² which reported a survivorship of 70% at 5 years and 60.4% at 10 years. The top three reasons for implant failure were aseptic loosening (5.2%), malalignment (1.7%), and deep infection (1.0%).

JOINT REGISTRIES

The advantages of joint registries in analyzing outcomes include analysis of large numbers of operative procedures, analysis of outcomes of a large number of surgeons with varied experience levels, the ability to study surgeon-, hospital-, and region-specific trends in practices, and uniform evaluation tools to facilitate comparisons.¹⁵,¹⁶,¹⁷ and ¹⁸

In 2007, 18 surgeons from 18 hospitals reported their results extracted from New Zealand’s National Joint Registry.¹⁶ Only two surgeons had performed more than 25 TAAs. New Zealand’s registry included 202 TAAs performed in 183 patients, with 7% failures at a mean follow-up of 28 months. Only two prostheses, one two-component fixed-bearing implant (58%) and one three-component mobile-bearing implant (22%), were used. Overall 5-year survival was 86%, with the majority of failures being due to aseptic loosening.

A larger registry from Sweden included 531 primary mobilebearing TAA implants (492 patients) reported between 1993 and 2005.¹⁸ Seventy-three percent of the TAAs had been performed by three surgeons in four hospitals. The 19% reported revision rate in this patient cohort is subject to a learning curve, with the busiest three surgeons noting an improvement in implant survivorship of 70% and 86% for the first 90 and second 132 TAAs, respectively. Most revision surgeries were due to the high number of technical errors and aseptic loosening, with nearly all cases of aseptic loosening occurring with the earlier design of the STAR prosthesis.

In a follow-up study of the Swedish Ankle Register, Henricson et al.¹⁹ reported on 780 prostheses implanted between 1993 and 2010. The overall 5-year survival rate was 81% and the 10-year survival rate was 69%. Interestingly, the 10-year survival rate was 72% for patients with rheumatoid arthritis (RA), 68% for patients with osteoarthritis (OA), and 66% for patients with posttraumatic arthritis. The survival rate was significantly lower for the single-coated STAR prosthesis implanted between 1993 and 1999. There was no significant difference in survival rates among the other prostheses (Ankle Evolutive System [AES], Buechel-Pappas, CCI, Hintegra, Mobility, and double-coated STAR). Aseptic loosening was the most common reason for revision and it was mainly a problem with the single-coated STAR.

The Norwegian Arthroplasty Register includes 257 primary TAAs in 245 patients (mean age 58 years) with an average follow-up of 4 years.¹⁷ Meaningful data from this database are for an early version of the STAR mobile-bearing prosthetic design (1996 to 2002) and a modern version of the same prosthesis (2000 to 2005). As noted by other authors, fewer revisions due to aseptic loosening of the tibial component were performed in the modern prostheses when compared to the earlier design.¹⁸^,²⁰^,²¹ The authors report a 5- and 10-year survivorship of 89% (modern) and 76% (early), respectively. These calculations include the outdated cemented two-component versions of a mobile-bearing prosthetic design that is no longer used. Altogether, 21 revisions were performed in 216 mobile-bearing implants. Six were for aseptic loosening; five occurred with the early design versus only one for the modern design. These authors defined revision as reoperation and did not designate removal of a metal component or conversion to arthrodesis as the end point for their overall survivorship analysis as is commonly reported in other studies.⁴^,²²,²³ and ²⁴ Recalculation of this registry’s survivorship without the six polyethylene exchanges and three other reoperations without removal of the metal components would make this study’s survivorship data consistent with those from other studies of modern TAA.

Outcomes of 515 primary mobile-bearing TAAs included in the Finnish Arthroplasty Register were published in 2010.¹⁵ Three centers each performed 100 TAAs or more, 4 centers performed between 10 and 50 TAAs, and 10 centers accounted for less than 10 TAAs each. Five-year survival was 83% with any reason for revision surgery as the end point and 95% isolating aseptic loosening as the end point. Their analysis of results failed to show a measurable difference in implant survival rates between high- and low-volume hospitals. Relative to the outcomes derived from the other Scandinavian national joint registries, the Finnish report suggests a high rate of ligamentous instability. Despite the majority of metal implants being well-fixed to bone, several TAAs showed ligamentous instability, which the authors attribute to the learning curve for TAA and persistent ligament attenuation secondary to preoperative deformity.

TAA RESULTS BASED ON PATIENT-SPECIFIC FACTORS

ETIOLOGY OF ANKLE ARTHRITIS

Multiple etiologies of ankle arthritis have been described. The majority of ankle arthritis is posttraumatic in nature. Posttraumatic ankle arthritis includes both ankle fracture and nonfracture traumatic conditions such as repetitive ankle sprains. In fact, studies indicate that approximately 51% to 80% of patients who underwent TAA had posttraumatic conditions
as the primary etiology of ankle arthritis.²⁵,²⁶ and ²⁷ Other common causes include inflammatory conditions such as RA and degenerative arthritis that cannot be attributed to any specific trauma. Less commonly, TAA has demonstrated efficacy for patients with gouty arthritis²⁸ and those with joint destruction secondary to hemochromatosis.²⁹ Results of TAA based on etiology are mixed. The skewed posttraumatic etiology of ankle arthritis and low sample sizes of RA and primary OA are the contributing factors.

According to some authors, results of TAA are less favorable and complication rates are higher in patients with posttraumatic arthritis than in those with OA and inflammatory arthritis.²⁵^,²⁷^,³⁰,³¹ and ³² In a 10-year follow-up study of 780 implants in the Swedish Ankle Register, Henricson et al.¹⁹ demonstrated a greater survival rate for patients with RA compared to OA and posttraumatic arthritis, although this was not significant. Rippstein et al.,²⁵ in a study of 233 TAAs, 123 of which were for posttraumatic arthritis and 36 of which were for RA, reported significantly less postoperative complications and reoperations in RA patients than in posttraumatic arthritis patients. Additionally, patients with RA experienced significantly greater improvement in AOFAS scores and reduction in pain compared to patients with posttraumatic arthritis. Alternatively, in a slightly smaller study of 158 TAAs with a subset of 12*7 performed for posttraumatic arthritis and 10 for RA, Giannini et al.²⁷ found no significant difference with respect to improvement in postoperative AOFAS scores between the two groups. On the contrary, one retrospective study demonstrated a slight, yet significantly greater, improvement in AOFAS ankle-hindfoot scores at final follow-up in patients with OA or posttraumatic arthritis compared to patients with RA.⁵ However, the small sample size of RA patients limits conclusions from these studies. Bai et al.³¹ compared the outcomes of 37 TAAs performed for posttraumatic arthritis and 30 TAAs performed for primary OA. They found no difference in AOFAS ankle-hindfoot scale scores, range of motion (ROM), or radiographic measurements at a mean follow-up of 38 months. One possible reason for these findings is that TAA for posttraumatic arthritis is frequently performed in a relatively younger patient population and patients who have undergone prior surgery to the ankle.¹⁷^,¹⁸^,²²^,³³ Finally, one joint registry and a meta-analysis suggest a trend of lessfavorable implant survivorship for inflammatory arthritis when compared to OA.⁷^,¹⁷

AGE

The mean age of patients receiving TAA consistently ranges from 50 to 60 years.⁴^,⁷^,⁸ Several investigations of TAA suggest that implant survivorship and functional outcomes are less favorable in younger patients.⁸^,¹⁷^,¹⁸^,³⁴ In a study of 303 TAAs in 303 patients, Spirt et al.³⁴ reported 5-year implant survivorship of 74% and 89% for patients under and over the age of 54, respectively. These authors also calculated that patients with a median age of 54 years or less had a 1.45-times greater risk for reoperation and a 2.65-times greater risk of implant failure than patients over the age of 54. Nonetheless, Kofoed and Lundberg-Jensen¹³ demonstrated, in a comparative study of a mobile-bearing prosthesis, including early-generation cemented and uncemented TAAs, that TAA survivorship was comparable for patients under and over 50 years of age. One TAA registry reported that a lower age at the time of TAA was associated with increased risk of revision.¹⁸ Data from 780 patients in the Swedish Ankle Register demonstrated that women younger than 60 years who suffered from OA or posttraumatic arthritis had a significantly higher risk of revision than patients older than 60 years.¹⁹

WEIGHT

Few studies report or analyze patient weight. One study reported on the body mass index (BMI) of 90 patients.³⁵ They found no significant change in mean BMI at 6 months, 1 year, 2 years, and 5 years postoperatively despite significant improvement in Ankle Osteoarthritis Scale scores and Short Form (SF)-36 Physical Component scores. The authors did not correlate BMI at the time of surgery and outcome scores or complications.

PREOPERATIVE DEFORMITY

One of the goals of TAA is to return the ankle to physiologic alignment. The magnitude of deformity that can be corrected at the time of TAA is unknown. Commonly, coronal plane deformity exceeding 10° to 15° is reported as a relative contraindication to TAA.¹¹^,²⁰^,²³^,³⁶,³⁷ and ³⁸ Doets et al.³⁸ reported an increased failure rate with a preoperative deformity of more than 10° in the coronal plane. On the contrary, Hobson et al.³⁹ compared patients with preoperative deformity of less than 10° to a group with coronal plane deformity of 11° to 30° and found no difference in ROM, complication rate, and survival rate. Sagittal plane deformity, typically with relative anterior translation of the talus to the tibia, may also result in persistent postoperative deformity, subluxation and edge loading of the polyethylene, osteolysis, and potential early implant failure.²³

THE SUBTALAR JOINT

Although necessary in select cases of TAA, subtalar arthrodesis may lead to diminished TAA survivorship in younger, higherdemand patients.²² Even with equal pain subscores, efficacy outcomes of TAA with subtalar arthrodesis will be inferior to those for isolated TAA, since outcome scoring for the ankle includes hindfoot ROM.⁴⁰^,⁴¹

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