583383

research-article2015

FAIXXX10.1177/1071100715583383Foot & Ankle InternationalBraito et al

Article

Effect of Coronal and Sagittal Alignment on Outcome After Mobile-Bearing Total Ankle Replacement

Foot & Ankle International® 1­–9 © The Author(s) 2015 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1071100715583383 fai.sagepub.com

Matthias Braito, MD1, Dietmar Dammerer, MD1, Andrea Reinthaler, MD1, Gerhard Kaufmann, MD1, Dennis Huber, MSc2, and Rainer Biedermann, MD1

Abstract Background: Comparably high revision rates are reported after total ankle replacement (TAR). Therefore, further critical analysis of the influence of implant position on clinical outcome is necessary. Methods: We analyzed the reliability and predictive value of previously published pre- and postoperative coronal and sagittal parameters in routine ankle radiographs on the clinical outcome of 84 HINTEGRA total ankle replacements (Newdeal, Lyon, France; Integra, Plainsboro, New Jersey). Mean follow-up was 4.0 years, and 15.5% of the TARs had revision surgery. Results: Mean postoperative American Orthopaedic Foot & Ankle Society ankle-hindfoot score was 71.3; mean postoperative verbal rating scale for pain was 3.1; and mean postoperative ankle range of motion was 26.4 degrees. Most tested radiologic parameters showed moderate or high intra- and interobserver reliability. With the numbers available, no significant difference in clinical outcome for all tested radiologic parameters could be detected. Conclusion: Our results indicate that mild malalignment of TAR, as assessed on routine ankle radiographs, did not affect midterm clinical outcome after TAR. Further multicenter studies with longer follow-up are needed to support our findings. Level of Evidence: Level III, comparative series. Keywords: total ankle replacement, radiologic analysis, clinical outcome, implant position Modern 3-component total ankle replacements (TARs) have become a reasonable treatment option for end-stage ankle osteoarthritis and have encouraging midterm clinical results.4,7,22,29,30 However, significantly lower survival rates than those after knee and hip replacements are reported, with an overall survival rate of 80% after 5 years and 70% after 10 years, respectively, in national registries.10,15,16 Common reasons for revision surgery are wound-healing problems, deep infections, prosthesis misalignment,2,6,8 incorrect soft tissue balancing,3 periprosthetic lucency with implant loosening,14,23 and periprosthetic ossification with reduced range of motion (ROM).18 The primary goals when implanting TAR are to restore a physiologic tibio-talo-calcaneal sagittal and coronal axis and to create a posterior tibial slope by aligning with an extramedullary guide. Furthermore, proper size and rotation of the components and ligamentous balance must be achieved to allow free motion of the polyethylene insert and to avoid high contact pressures.8 For further improvements of prosthetic design and technique of implantation, radiologic analysis of implant position with regard to its correlation with clinical outcome and revision rate is needed. Barg et al introduced the anteroposterior offset ratio as a useful

predictor for postoperative pain and function after TAR.2 Cenni et al investigated the influence of component position on postoperative ROM and found a significantly higher ROM with the talar component located and inclined more anteriorly than the tibial.5 Fluoroscopy was used on all postoperative radiographic examinations to ensure standardized and true anteroposterior and lateral views of both components. In routine radiographic follow-up examinations, this is usually not performed, because it exposes the patient to a risk of higher radiation burden and it is time-consuming.

1

Department of Orthopaedics, Innsbruck Medical University, Innsbruck, Austria 2 Department of Experimental Orthopaedics, Innsbruck Medical University, Innsbruck, Austria Investigation performed at the Department of Orthopaedic Surgery, Medical University of Innsbruck, Austria Corresponding Author: R. Biedermann, MD, Associate Professor, Consultant Orthopaedic Surgeon, Department of Orthopaedics, Innsbruck Medical University, Anichstrasse 35, A–6020 Innsbruck, Austria. Email: [email protected]

Downloaded from fai.sagepub.com at TUFTS UNIV on December 1, 2015

2

Foot & Ankle International 

We therefore aimed to analyze the prospective value of routine radiographic alignment measurement of a modern 3-component nonconstrained prosthesis on clinical outcome and revision rate. We hypothesized that sagittal and coronal malalignment above a threshold value would lead to worse clinical outcome and earlier and higher revision rate, respectively.

Methods All patients with TAR for primary, posttraumatic, or inflammatory ankle osteoarthritis since the introduction of the HINTEGRA total ankle prosthesis (Newdeal, Lyon, France; Integra, Plainsboro, NJ, USA; Figure 1) in our institution in January 2004 and with a follow-up of more than 12 months were included in the study. All procedures were performed by 2 experienced orthopaedic surgeons who specialized in foot and ankle surgery. Routine follow-up examinations included anteroposterior and lateral ankle radiographs and clinical examinations annually. In a prospectively planned follow-up examination, patients were also invited for gait analysis and completion of American Orthopaedic Foot and Ankle Society (AOFAS) and Foot and Ankle Outcome Score (FAOS) questionnaires.13 The local ethical board approved the study, and all participating patients signed informed consent forms before data collection.

Study Population Between January 2004 and December 2012, a consecutive series of 96 patients (99 TARs) underwent TAR using HINTEGRA at our institution. Fifteen patients (15 TARs) were excluded from this study: 6 patients were lost to follow-up; 4 patients had revision surgery within 12 months after implantation for septic complications; 3 patients had a conversion to TAR after ankle fusion; 1 patient had TAR within the last 12 months; and 1 patient had died within 1 year after the operation. Thus, the patient population for radiologic and clinical analysis consisted of 81 patients (84 TARs; 38 male, 43 female) at a mean age of 61.7 ± 12.6 years. AOFAS ankle-hindfoot and FAOS scores were obtained from 59 patients (61 ankles); 53 operations were performed owing to posttraumatic osteoarthritis, 16 for inflammatory osteoarthritis, and 15 because of primary osteoarthritis. Mean follow-up was 4.0 ± 2.0 years.

Clinical Analysis Clinical analysis included ROM measurement with a goniometer placed along the lateral border of the leg and foot and assessment of subjective outcome with the AOFAS ankle-hindfoot score, FAOS, and verbal rating scale (VRS; 0-10) for pain. Clinical investigations and a detailed retrospective analysis of medical records with respect to revision

Figure 1.  The HINTEGRA total ankle prosthesis (Newdeal, Lyon, France; Integra, Plainsboro, New Jersey). Photo material is copyrighted and all rights reserved to Smith & Nephew GmbH, Marl, Germany.

surgery were performed by 2 independent reviewers, who did not perform an operation on any of the patients.

Radiologic Analysis Radiologic analysis was performed twice with Ico View (ITH Icoserve Technology for Healthcare, GmbH, Innsbruck, Austria) at 2 different time points by 2 investigators, who were blinded to the clinical outcome. Radiographs taken before surgery (preoperative), on the postoperative day (postoperative 1), and at the latest follow-up (postoperative 2) were reviewed. In case of revision, the last postoperative radiograph before revision surgery was used. Anteroposterior and lateral ankle radiographs were routinely taken with the patient in a standing, weightbearing position without the use of fluoroscopy. Moderate or high intra- and interobserver reliability of the chosen radiologic measurement methods has already been shown, and normal values have been suggested.2,20,25,26 On the anteroposterior view, the anatomic lateral distal tibial angle (normal value: 85 to 95 degrees) was measured. The tibiotalar angle (normal value: –5 to 5 degrees) was defined as the angle between the longitudinal axis of the tibia and a perpendicular line to the superior articular surface of the talus. On the lateral view, the anatomic anterior distal tibial angle (normal value: 80 to 90 degrees) showed the tibial slope. Sagittal tibiotalar alignment was quantified by 3 methods: The tibial axis–talus ratio (tibial-talar ratio; normal value: 27% to 42%) described by Tochigi et al,25,26

Downloaded from fai.sagepub.com at TUFTS UNIV on December 1, 2015

3

Braito et al the anteroposterior offset ratio (normal value: 0) according to Barg et al,2 and the contact point ratio (normal value: 40% to 45%) described by Hintermann et al.11 Figures 2-12 show the radiographic measurements via line drawings.

Statistics Statistical analysis was performed with the SPSS 20.0 statistical software package (IBM Corporation, Armonk, New York, USA). Quantitative, normally distributed variables are described as means and standard deviations. Quantitative variables from nonnormal distributions are described by medians and interquartile ranges. All quantitative variables were tested with a Kolmogorov-Smirnov test for normal distribution. Intra- and interobserver variability was tested with Pearson correlation. The Student t test for paired samples was used to test for mean value differences between pre- and postoperative clinical outcome parameters. The χ2 test was performed to determine whether or not revision surgery was related to radiologic outcome. To test for predictive value of measured radiologic parameters, patients were divided into 2 groups (inside vs outside the normal value) based on suggested normal values in the literature. The Student t test for independent samples was used to test for mean value differences between both groups. A post hoc sample size calculation (alpha value: 0.05) showed low to moderate statistical power (beta value: 0.21 to 0.59). We chose a significance level of P = .05 for all tests.

Results Revision Rate According to the definitions of Henricson et al,9 revision surgery was necessary in 15.5% of cases (n = 13) after an average of 3.1 ± 1.3 years. Revision surgeries consisted of 11 exchanges of 1 or more of the prosthetic components (due to septic complications, aseptic loosening, osteolysis, ossifications, impingement, fracture of the inlay), 1 amputation for septic complications, and 1 conversion to ankle arthrodesis because of instability.

Clinical Outcome VRS, AOFAS, and FAOS scores could be obtained from 59 patients (61 ankles), whereas ROM measurements were available for all patients. Means and standard deviations of all measured clinical parameters are shown in Table 1. All parameters except ankle ROM improved significantly after TAR.

Radiologic Outcome Means and standard deviations of all measured radiologic parameters and normal values as suggested in the literature

Figure 2.  Preoperative anatomic lateral distal tibial angle measurement (normal value: 85 to 95 degrees).

are shown in Table 2. Most tested radiologic parameters showed moderate or high intra- and interobserver reliability (R = 0.41 to 0.92, P < .001). With the numbers available, the tested radiologic parameters showed no significant influence on revision rate (P = .129 to 1.000) or clinical outcome (T = –2.494 to 1.882, P = .056 to .989).

Discussion The overall revision rate in our study cohort, based on the definition of Henricson et al,9 was 15.5% after a mean follow-up of 4.0 years, comparable to the numbers reported in the Swedish ankle registry10 but higher than the revision rate at the designers institution.4,15,16 Satisfactory clinical outcome was achieved for most patients; mean total AOFAS score was 71.3, indicating fair results comparable to those in the present literature.4,30 All FAOS subscales (pain, symptoms, activities of daily living, sports, and quality of life) and VRS for pain improved significantly after the procedure. We found no significant difference between pre- and postoperative ankle ROM (27.3 vs 26.4 degrees). Patients with a higher preoperative joint mobility lost motion after TAR, whereas patients with lower preoperative ROM gained motion. These findings are similar to Ajis et al,1 who found

Downloaded from fai.sagepub.com at TUFTS UNIV on December 1, 2015

4

Foot & Ankle International 

Figure 4.  Preoperative tibiotalar angle measurement (normal value: –5 to 5 degrees).

Figure 3.  Postoperative anatomic lateral distal tibial angle measurement (normal value: 85 to 95 degrees).

a mean ROM of 22.7 degrees preoperatively and 24.3 degrees 12 months postoperatively in 3 types of TAR. Radiographic analysis showed satisfactory component positioning in most cases. One tibial component had a valgus position (anatomic lateral distal tibial angle: 83 degrees), while 4 tibial components had a varus position between 96 and 98 degrees. In contrast to Pyevich at al, who found a significantly higher rate of pain with the tibial component’s position in more than 4 degrees of valgus,20 no significant effect of the anatomic lateral distal tibial angle on overall clinical outcome was noted in our population. Other authors stated that a varus position of the tibial component may lead to medial translation force of the talar component, resulting in higher incidence of medial malleolar pain syndrome.3 While we were not able to detect higher pain scores in ankles with varus position of the tibial component, 2 of the 4 ankles with varus position of the tibial component of more than 95 degrees were revised for aseptic loosening. Furthermore, even when preoperative tibial angle exceeded 10 degrees of varus or valgus (4 cases), a neutral

postoperative alignment could be achieved in all cases. However, 1 of those 4 prostheses had to be revised for mechanical failure (osteolysis). Therefore, our findings support existing literature that TAR can also be successfully performed with severe preoperative deformity, when congruency and ligamentous balance are achieved postoperatively.12,21,27,28 We therefore believe that coronal malalignment within a safe zone, which has not been established yet, in an otherwise well-balanced hindfoot does not lead to a poor clinical outcome or higher revision rates.17 Sagittal positioning of the talar component is thought to be a critical aspect in the operative technique of TAR. Especially anterior malpositioning in the sagittal plane was correlated with decreased ROM due to polyethylene bearing lift-off after STAR (Scandinavian Total Ankle Replacement) under in vitro conditions.24 Furthermore, anteroposterior implant malpositioning after TAR has been considered a possible cause of premature implant failure.24-26 However, we were not able to detect an influence of sagittal positioning on ankle ROM or implant failure rate. Anterior translation of the talus with respect to the tibia was previously assessed using different measurements, such as the anteroposterior offset ratio,2 the tibial-talar ratio,19,25,26

Downloaded from fai.sagepub.com at TUFTS UNIV on December 1, 2015

5

Braito et al

Figure 6.  Preoperative anatomic anterior distal tibial angle measurement (normal value: 80 to 90 degrees).

Figure 5.  Postoperative tibiotalar angle measurement (normal value: –5 to 5 degrees).

and the contact point ratio.11 In our study, the highest intraand interobserver reliability was found for the tibial-talar ratio, which averaged 32.5% ± 5.4% and showed no influence on clinical outcome. Using the anterior-posterior offset ratio of Barg et al,2 we found no significant differences among the anteroposterior offset ratio subgroups (ie, less than 0, 0, more than 0) with regard to postoperative AOFAS score and ankle ROM. The position of the contact point between the tibial and talar component11 was 43.5% ± 9.4% in our population and showed no significant influence on clinical outcome. Moreover, the measurement of the contact point was prone to higher interobserver variability, which might partially be explained by the lack of true lateral views. It was also somewhat surprising that no statistically significant association was observed between postoperative ankle ROM and the tibial slope or the anterior translation of the talar component. Our findings differ from data reported by Cenni et al,5 who found a significantly larger ROM in ankles with the talar component located and inclined more anteriorly than the tibial component. We assume that this

Figure 7.  Postoperative anatomic anterior distal tibial angle measurement (normal value: 80 to 90 degrees).

Downloaded from fai.sagepub.com at TUFTS UNIV on December 1, 2015

6

Foot & Ankle International 

Figure 8.  Preoperative tibial axis–talus ratio measurement (tibiotalar ratio; normal value: 27% to 42%).

Figure 9.  Postoperative tibial axis–talus ratio measurement (tibiotalar ratio; normal value: 27% to 42%).

Figure 10.  Preoperative anteroposterior offset ratio measurement (normal value: 0).

Figure 11.  Postoperative anteroposterior offset ratio measurement (normal value: 0).

Downloaded from fai.sagepub.com at TUFTS UNIV on December 1, 2015

7

Braito et al Table 1. (continued)

ROM  Preoperative  Postoperative

Mean ± SD

P

27.3 ± 13.3 26.4 ± 9.4

   .341  

Abbreviations: AOFAS, American Orthopaedic Foot and Ankle Society; FAOS, Foot and Ankle Outcome Score; QOL, quality of life; ROM, range of motion; VRS, Verbal Rating Scale. a The Student t test for paired samples was used to test for mean value differences between pre- and postoperative clinical outcome parameters.

Table 2.  Means and Standard Deviations of All Measured Radiologic Parameters and Normal Values as Suggested in the Literature. Mean ± SD

Figure 12.  Postoperative contact point ratio measurement (normal value: 40% to 45%). Table 1.  Means and Standard Deviations of All Measured Clinical Parameters. Mean ± SD VRS pain  Preoperative  Postoperative AOFAS global  Preoperative  Postoperative FAOS subscales  Pain   Preoperative   Postoperative  Symptoms   Preoperative   Postoperative  ADL   Preoperative   Postoperative  Sports   Preoperative   Postoperative  QOL   Preoperative   Postoperative

8.3 ± 1.4 3.1 ± 2.2 — 71.3 ± 18.8

28.2 ± 13.6 70.4 ± 18.8 33.5 ± 23.2 63.7 ± 22.0 36.8 ± 20.1 72.6 ± 21.2 17.2 ± 15.4 45.8 ± 24.2 13.4 ± 11.5 43.5 ± 24.5

P  

Effect of Coronal and Sagittal Alignment on Outcome After Mobile-Bearing Total Ankle Replacement.

Comparably high revision rates are reported after total ankle replacement (TAR). Therefore, further critical analysis of the influence of implant posi...
732KB Sizes 0 Downloads 9 Views