Severe Ankle Varus Malalignment Management with a Fix-Bearing Total Ankle Replacement Implant

Category: Ankle Arthritis; Ankle Introduction/Purpose: Despite Total ankle replacement (TAR) is currently considered a safe procedure and an optimal alternative to ankle fusion in end-stage ankle arthritis, still the precise indications and the limits of TAR are a highly controversial topic. Coronal malalignment correction with TAR is also a widely debated point. Purpose of this work was to confront two patient groups undergoing Total Ankle Replacement (a group of patients with a pre-operative ankle varus misalignment on coronal plane, versus a group of patients with neutral pre-operative ankle alignment) comparing clinical and radiographic results, complications and survival rate of the implant. Methods: 171 patients, all affected by end-stage ankle arthritis and all treated with total ankle replacement with fix-bearing Trabecular Metal Total Ankle System (Zimmer-Biomet, Warsaw IN, USA), were divided in two groups. Group ‘A’ included 159 patients having a normal alignment or a slight valgus or valgus misalignment within 10°. Group ‘B’ included 12 patients with a pre- operative varus malalignment >10°. Each patient of the study was evaluated, clinically and radiologically, before and after surgery, and subsequently at 6 and 12 months follow-up, then once every year. Clinical evaluation comprised some score: visual analogue scale (VAS); American Foot and Ankle Society Score (AOFAS); and the Short Form (SF)-12 Quality of Life, in both of its ‘physical’ (PCS) and ‘mental’ (MCS) components. Radiological evaluated parameters were: tibio-talar surface angle (TTS), lateral distal tibial angle (LDTA-α), and anterior distal tibial angle (ADTA-β). Complications and implant survival rate were evaluated for both groups. Results: At a mean follow up of 2,75 years (range 2.09-5.66 years), improvement was found in all clinical and radiological evaluated parameters of two groups, without statistically significant differences between two groups at last follow-up. In group A most frequent reported complication was the presence of a fibular symptomatic hardware. Major complication were a deep arthroplasty infection and a 3 months post-operative ankle fracture-dislocation/periprostethic fracture. In group B there were 2 delayed wound healing, and one case of loosening of correction. No significant difference in complication and revision rate comparison between two groups was founded. Conclusion: Mid-term results reported in the present show that in patients with varus greater than 10 ° it is possible to obtain a post-operative alignment similar to pre-operative neutral ankles and that the complication rate is not greater. Despite this, is Authors opinion that major and complex misalignment should be approached by skilled surgeons, with an adequate learning curve experience, with a wide range of cases and in highly specialized centers.

Download Full-text

Survival of the Scandinavian Total Ankle Replacement (STAR)

Foot & Ankle Orthopaedics ◽

10.1177/2473011418s00223 ◽

2018 ◽

Vol 3 (3) ◽

pp. 2473011418S0022

Author(s):

Arno Frigg ◽

Ursula Germann ◽

Martin Huber ◽

Monika Horisberger

Keyword(s):

Survival Rate ◽

Total Ankle Replacement ◽

Ankle Arthritis ◽

Ankle Osteoarthritis ◽

Metallic Component ◽

Kaplan Meier ◽

Ankle Replacement ◽

One Year ◽

End Stage

Category: Ankle Arthritis Introduction/Purpose: The purpose of this study was to evaluate survival and clinical outcome of the Scandinavian total ankle replacement (STAR) prosthesis after a minimum of ten years up to a maximum of 19 years. Methods: Fifty STAR prostheses in 46 patients with end stage ankle osteoarthritis operated between 1996 and 2006 by the same surgeon (MH) were included. Minimal follow-up was ten years (median 14.6 years, 95% confidence interval [CI] 12.9-16.4). Clinical (Kofoed score) and radiological assessments were taken before the operation and at one, ten (+2), and 16 (±3) years after implantation. The primary endpoint was defined as exchange of the whole prosthesis or conversion to arthrodesis (def. 1), exchange of at least one metallic component (def. 2), or exchange of any component including the inlay (due to breakage or wear) (def. 3). Survival was estimated according to Kaplan-Meier. Further reoperations related to STAR were also recorded. Results: The ten year survival rate was (def. 1) 94% (CI 82- 98%), (def. 2) 90% (CI, 77-96%), and (def. 3) 78% (CI 64-87%). The 19-year survival rate was (def. 1) 91% (CI 78-97%), (def. 2) 75% (CI 53-88%), and (def. 3) 55% (CI 34-71%). Considering any re-operations related to STAR, 52% (26/50) of prostheses were affected by re-operations. Mean pre-operative Kofoed score was 49, which improved to 84 after one year (n = 50), to 90 after ten years (n = 46), and to 89 after 16 years (n = 28). Conclusion: The survival rate for def. 1 and 2 was high. However, re-operations occurred in 52% of all STAR prosthesis.

Download Full-text

Total Ankle Arthroplasty Survivorship, Complication, and Revision Rates in Patients Younger Than 55 Years

Foot & Ankle Specialist ◽

10.1177/1938640020980925 ◽

2021 ◽

pp. 193864002098092

Author(s):

Devon W. Consul ◽

Anson Chu ◽

Travis M. Langan ◽

Christopher F. Hyer ◽

Gregory Berlet

Keyword(s):

Case Series ◽

Ankle Arthrodesis ◽

Total Ankle Replacement ◽

Implant Survival ◽

Ankle Arthritis ◽

Single Institution ◽

Retrospective Case ◽

Ankle Replacement ◽

End Stage

Total ankle replacement has become a viable alternative to ankle arthrodesis in the surgical management of advanced ankle arthritis. Total ankle replacement has generally been reserved for patients who are older and for those who will have a lower demand on the replacement. The purpose of the current study is to review patient outcomes, complications, and implant survival in patients younger than 55 years who underwent total ankle replacement at a single institution. A single-center chart and radiographic review was performed of consecutive patients who underwent total ankle replacement for treatment of end-stage ankle arthritis. All surgeries were performed by 1 of 5 fellowship-trained foot and ankle surgeons at a single institution. A total of 51 patients met inclusion criteria with a mean follow-up of 31.2 months (SD = 16.2). Implant survival was 94%, There were 7 major complications (13%) requiring an unplanned return to the operating room and 8 minor complications (15%) that resolved with conservative care. The results of this study show that total ankle replacement is a viable treatment option for patients younger than 55 years. Levels of Evidence: A retrospective case series

Download Full-text

Prospective Randomized Trial Comparing Mobile-bearing and Fixed-bearing Total Ankle Replacement

Foot & Ankle Orthopaedics ◽

10.1177/2473011418s00093 ◽

2018 ◽

Vol 3 (3) ◽

pp. 2473011418S0009

Author(s):

James Nunley ◽

Samuel Adams ◽

James DeOrio ◽

Mark Easley

Keyword(s):

Clinical Outcomes ◽

Mobile Bearing ◽

Cyst Formation ◽

Total Ankle Replacement ◽

Fixed Bearing ◽

Ankle Arthritis ◽

Ankle Replacement ◽

Significant Difference ◽

End Stage

Category: Ankle Arthritis Introduction/Purpose: Outcomes of total ankle replacement for the treatment of end-stage ankle arthritis continue to improve. Debate continues whether a mobile-bearing total ankle replacement (MB-TAR) or a fixed-bearing total ankle replacement (FB-TAR) is superior, with successful outcomes reported longterm for MB-TAR and at intermediate-to-longterm follow-up for newer generation FB-TAR. Although comparisons between the two total ankle designs have been reported, to our knowledge, no investigation has compared the two designs with a high level of evidence. This prospective, randomized controlled trial conducted at a single institution compares patient satisfaction, functional outcomes and radiographic results of the mobile-bearing STAR and the fixed-bearing Salto-Talaris in the treatment of end-stage ankle arthritis. Methods: This investigation was approved by our institution’s IRB committee. Between November 2011 and November 2014, adult patients with end-stage ankle osteoarthritis failing nonoperative treatment were introduced to the study. With informed consent, 100 patients (31 male and 69 female, average age 65, range 35 to 85) were enrolled; demographic comparison between the two cohorts was similar. Exclusion criteria included inflammatory arthropathy, neuropathy, weight exceeding 250 pounds, radiographic coronal plane deformity greater than 15 degrees or extensive talar dome wear pattern (“flat top talus”). Prospective patient-reported outcomes, physical exam and standardized weightbearing ankle radiographs were obtained preoperatively, at 6 and 12 months postoperatively, and then at yearly intervals. Data collection included visual analog pain score (VAS), short form 36 (SF-36), foot and ankle disability index (FADI), short musculoskeletal functional assessment (SMFA) and AOFAS ankle-hindfoot score. Surgeries were performed by non-design team orthopaedic foot and ankle specialists with total ankle replacement expertise. Statistically analysis was performed by a qualified statistician. Results: At average follow-up of 4.5 years (range 2-6 years) complete clinical data and radiographs were available for 84 patients; 7 had incomplete data, one had died, 4 were withdrawn after enrolling but prior to surgery and 4 were lost to follow-up. In all outcome measures, the entire cohort demonstrated statistically significant improvements from preoperative evaluation to most recent follow-up. There was no statistically significant difference in improvement in clinical outcomes between the two groups. Radiographically, tibial lucency/cyst formation was 26.8% and 20.9% for MB-TAR and FB-TAR, respectively. Tibial settling/subsidence occurred in 7.3% of MB-TAR. Talar lucency/cyst formation occurred in 24.3% and 2.0% of MB-TAR and FB-TAR, respectively. Talar subsidence was observed in 21.9% and 2.0% of MB-TAR and FB-TAR, respectively. Re-operations were performed in 8 MB-TAR and 3 FB-TAR, with the majority of procedures being to relieve impingement or treat cysts and not to revise or remove metal implants. Conclusion: For the first time, with a high level of evidence, our study confirms that patient reported and clinical outcomes are favorable for both designs and that there is no significant difference in clinical improvement between the two implants. The incidence of lucency/cyst formation was similar for MB-TAR and FB-TAR for the tibial component, but the MB-TAR had greater talar lucency/cyst formation and tibial and talar subsidence. As has been suggested in previous studies, clinical outcomes do not correlate with radiographic findings. Re-operations were more common for MB-TAR and in the majority of cases were to relieve impingement or treat cysts rather than revise or remove metal implants.

Download Full-text

Outcomes of Revision Total Ankle Replacement: A Minimum of 2 Year Follow-up

Foot & Ankle Orthopaedics ◽

10.1177/2473011419s00320 ◽

2019 ◽

Vol 4 (4) ◽

pp. 2473011419S0032

Author(s):

Andrea Pujol Nicolas ◽

Jayasree Ramas Ramaskandhan ◽

Triin Nurm ◽

Malik Siddique

Keyword(s):

Total Ankle Replacement ◽

Return To Sports ◽

Ankle Arthritis ◽

Ankle Replacement ◽

Patient Reported ◽

The Mean ◽

End Stage ◽

Average Domain

Category: Ankle, Ankle Arthritis Introduction/Purpose: Total ankle replacement as a valid treatment for end stage ankle arthritis, is gaining popularity and every year there is an increasing number of procedures. With revision rates as high as 21% at 5 years and 43% at 10 years there is a need for understanding and reporting the outcome of revision ankle replacement. Our aim was to study the patient reported outcomes following revision TAR with a minimum of 2 year follow up. Methods: All patients that underwent a revision total ankle replacement between 2012 and 2016 were included in the study. All patients received a post-operative questionnaire comprising of MOX-FQ score, EQ-5D (UK) and Foot and Ankle outcomes scores (FAOS) and patients satisfaction questionnaire with a minimum of 2 years follow up. Results: 32 patients had a revision total ankle replacement between 2012 and 2016. 2 patients were deceased therefore 30 patients were included in the study. 5 patients declined participation for completing questionnaires. We received 21 (66%) completed questionnaires. The mean MOX-FQ average domain score for pain was 58.8, walking/standing 65.8 and social function was 48.2. The mean FAOS scores were 50.7 for pain, 50.6 for symptoms, 54.9 for ADL and 28.2 for quality of life. The mean overall health score today for EQ-5D was 67.8/100. 45% of patients were satisfied with the pain relief and return to sports and recreation obtained following the operation, 48% were satisfied with the improved in daily activities. 52% were overall satisfied with the results from surgery. Conclusion: Revision total ankle replacement gives overall satisfactory results demonstrated from patients reported outcomes at a minimum of 2 years following surgery.

Download Full-text

An Update on the Canadian Experience Treating End Stage Ankle Arthritis with Fusion or Replacement Surgery

Foot & Ankle Orthopaedics ◽

10.1177/2473011418s00130 ◽

2018 ◽

Vol 3 (3) ◽

pp. 2473011418S0013

Author(s):

Kevin Wing ◽

Jason Sutherland ◽

Timothy Daniels ◽

Peter Dryden ◽

Murray Penner ◽

...

Keyword(s):

Patient Characteristics ◽

Total Ankle Replacement ◽

Ankle Arthritis ◽

Ankle Fusion ◽

Ankle Replacement ◽

Sf 36 ◽

Patient Reported ◽

The Mean ◽

End Stage

Category: Ankle Arthritis Introduction/Purpose: There is much interest in the surgical treatment of end-stage ankle arthritis. Our team has previously reported comparisons between ankle fusion and replacement outcomes for 321 ankles at 5.5 years from a prospective cohort. This research extends the observational period for another six years, ending in 2013. Our primary hypothesis is that patient-reported clinical outcomes for ankle fusion and replacements would be similar at last follow up. Methods: Patients in the Canadian Orthopaedic Foot and Ankle Society (COFAS) Prospective Ankle Reconstructive Database were treated with total ankle replacement (involving Agilty, Star, Mobilty, Hintegra, or Inbone) or fusion (open or arthroscopic). Patient characteristics collected included demographics, comorbidities, smoking status and body mass index. Patient-reported outcomes (PROs) completed by patients were the Ankle Osteoarthritis Scale (AOS) and Short Form-36 (SF-36). Pre-operative and most recent patient data, with at least four years follow-up, were analyzed. Sensitivity analyses excluded ankles that had undergone revision. A linear mixed-effects regression model compared scores between total ankle replacement and fusion groups, adjusting for patient characteristics, baseline data and surgeon. Results: The sample included 844 ankles (556 ankle replacements and 284 arthrodesis). The mean follow up period was 8.0 years (standard deviation 3.1 years), with minimum and maximum of 4 and 14 years, respectively. Patients treated with arthrodesis were younger, more likely to be diabetic and smokers, and somewhat less likely to have inflammatory arthritis. Overall, 19.4% of ankle fusion and 30.8% of ankle replacements underwent all-cause re-operation. The mean AOS total score improved from 58.6 points pre-operatively to 31.4 post-operatively (delta 27.2), and from 57.0 to 26.9 points (delta 30.1) in the ankle replacement group. Differences in the change in AOS and SF-36 scores between the arthrodesis and ankle replacement groups were minimal after adjustment for baseline characteristics and surgeon. Conclusion: Clinical outcomes of total ankle replacement and ankle arthrodesis were comparable in a diverse cohort of patients whose follow up period ranged between 4 and 14 years post-operatively.

Download Full-text

Clinical Radiographical Outcomes and Complications after a Brand-New Total Ankle Replacement Design through an Anterior Approach: A Retrospective at a Short-Term Follow Up

Journal of Clinical Medicine ◽

10.3390/jcm10112258 ◽

2021 ◽

Vol 10 (11) ◽

pp. 2258

Author(s):

Massimiliano Mosca ◽

Silvio Caravelli ◽

Emanuele Vocale ◽

Simone Massimi ◽

Davide Censoni ◽

...

Keyword(s):

Mean Value ◽

Total Ankle Replacement ◽

P Value ◽

Ankle Osteoarthritis ◽

Ankle Replacement ◽

Patient Reported ◽

End Stage ◽

Term Performance ◽

Replacement Design

Recently, the progress in techniques and in projecting new prosthetic designs has allowed increasing indications for total ankle replacement (TAR) as treatment for ankle osteoarthritis. This retrospective work comprehended 39 subjects aged between 47 and 79 years old. The patients, observed for at least 12 months (mean follow up of 18.2 ± 4.1 months), have been evaluated according to clinical and radiological parameters, both pre- and post-operatively. The AOFAS and VAS score significantly improved, respectively, from 46.2 ± 4.8 to 93.9 ± 4.1 and from 7.1 ± 1.1 to 0.7 ± 0.5 (p value < 0.05). At the final evaluation, the mean plantarflexion passed from 12.2° ± 2.3° to 18.1° ± 2.4° (p value < 0.05) and dorsiflexion from a pre-operative mean value of 8.7° ± 4.1° to 21.7° ± 5.4° post-operatively (p value < 0.05). This study found that this new total ankle replacement design is a safe and effective procedure for patients effected by end-stage ankle osteoarthritis. Improvements have been demonstrated in terms of range of motion, radiographic parameters and patient-reported outcomes. However, further studies are needed to assess the long-term performance of these prostheses.

Download Full-text

Prospective Randomized Trial Comparing Mobile-Bearing and Fixed-Bearing Total Ankle Replacement

Foot & Ankle International ◽

10.1177/1071100719879680 ◽

2019 ◽

Vol 40 (11) ◽

pp. 1239-1248 ◽

Cited By ~ 4

Author(s):

James A. Nunley ◽

Samuel B. Adams ◽

Mark E. Easley ◽

James K. DeOrio

Keyword(s):

Mobile Bearing ◽

Cyst Formation ◽

Total Ankle Replacement ◽

Fixed Bearing ◽

Level Of Evidence ◽

Ankle Replacement ◽

Significant Difference ◽

Patient Reported ◽

End Stage

Background: Outcomes of total ankle replacement for the treatment of end-stage ankle arthritis continue to improve. Debate continues whether a mobile-bearing total ankle replacement (MB-TAR) or a fixed-bearing total ankle replacement (FB-TAR) is superior, with successful outcomes reported long term for MB-TAR and at intermediate- to long-term follow-up for newer generation FB-TAR. Although comparisons between the 2 total ankle designs have been reported, to our knowledge, no investigation has compared the 2 designs with a high level of evidence. This prospective, randomized controlled trial conducted at a single institution compares patient satisfaction, functional outcomes, and radiographic results of the mobile-bearing STAR and the fixed-bearing Salto-Talaris in the treatment of end-stage ankle arthritis. Methods: Between November 2011 and November 2014, adult patients with end-stage ankle osteoarthritis failing nonoperative treatment were introduced to the study. With informed consent, 100 patients (31 male and 69 female, average age 65 years, range 35-85 years) were enrolled; a demographic comparison between the 2 cohorts was similar. Exclusion criteria included inflammatory arthropathy, neuropathy, weight exceeding 250 pounds, radiographic coronal plane deformity greater than 15 degrees, or extensive talar dome wear pattern (“flat-top talus”). Prospective patient-reported outcomes, physical examination, and standardized weightbearing ankle radiographs were obtained preoperatively, at 6 and 12 months postoperatively, and then at yearly intervals. Data collection included visual analog pain score, Short Form 36, Foot and Ankle Disability Index, Short Musculoskeletal Functional Assessment, and American Orthopaedic Foot & Ankle Society ankle-hindfoot score. Surgeries were performed by a nondesign team of orthopedic foot and ankle specialists with total ankle replacement expertise. Statistical analysis was performed by a qualified statistician. At average follow-up of 4.5 years (range, 2-6 years) complete clinical data and radiographs were available for 84 patients; 7 had incomplete data, 1 had died, 4 were withdrawn after enrolling but prior to surgery, and 4 were lost to follow-up. Results: In all outcome measures, the entire cohort demonstrated statistically significant improvements from preoperative evaluation to most recent follow-up with no statistically significant difference between the 2 groups. Radiographically, tibial lucency/cyst formation was 26.8% and 20.9% for MB-TAR and FB-TAR, respectively. Tibial settling/subsidence occurred in 7.3% of MB-TAR. Talar lucency/cyst formation occurred in 24.3% and 2.0% of MB-TAR and FB-TAR, respectively. Talar subsidence was observed in 21.9% and 2.0% of MB-TAR and FH-TAR, respectively. Reoperations were performed in 8 MB-TARs and 3 FH-TARs, with the majority of procedures being to relieve impingement or treat cysts and not to revise or remove metal implants. Conclusion: With a high level of evidence, our study found that patient-reported and clinical outcomes were favorable for both designs and that there was no significant difference in clinical improvement between the 2 implants. The incidence of lucency/cyst formation was similar for MB-TAR and FH-TAR for the tibial component, but the MB-TAR had greater talar lucency/cyst formation and tibial and talar subsidence. As has been suggested in previous studies, clinical outcomes do not necessarily correlate with radiographic findings. Reoperations were more common for MB-TAR and, in most cases, were to relieve impingement or treat cysts rather than revise or remove metal implants. Level of Evidence: Level I, prospective randomized study.

Download Full-text

Perspectives in Treatments of End-Stage Ankle Arthritis Among Orthopaedic Surgeons

Foot & Ankle Orthopaedics ◽

10.1177/2473011418779976 ◽

2018 ◽

Vol 3 (3) ◽

pp. 247301141877997

Author(s):

Phinit Phisitkul ◽

Natalie Glass ◽

Patrick B. Ebeling ◽

Sandra E. Klein ◽

Jeffrey E. Johnson

Keyword(s):

United States ◽

The United States ◽

Total Ankle Replacement ◽

Ankle Arthritis ◽

Cross Sectional Survey ◽

Level Of Evidence ◽

Cross Sectional ◽

Orthopaedic Surgeons ◽

Ankle Replacement ◽

End Stage

Background: This study aimed to assess the preferred operative treatment for patients over the age of 60 with end-stage ankle arthritis and perspectives on total ankle replacement (TAR) among American Orthopaedic Foot & Ankle Society (AOFAS) members. Associated factors were analyzed for potential contraindications among members with different levels of experience. Method: A questionnaire containing 6 questions was designed and sent to 2056 members of the AOFAS. Responses were received from 467 orthopaedic surgeons practicing in the United States (76%), Canada (5%), and 26 other countries (20%). Participants were grouped for response comparisons according to country as well as experience level. Differences in contraindications were compared using χ2 tests or exact tests. Results: Respondents practicing in the United States and surgeons who perform 11 or more TARs per year tended to recommend operative treatments favoring TAR and displayed recognition of its increasing role ( P < .05). Overall, respondents felt that 41% of typical patients over 60 years old with end-stage arthritis would be best treated with TAR. Talus avascular necrosis, morbid obesity (body mass index >40 kg/m2), and poorly controlled diabetes with neuropathy were most recognized as the absolute contraindications to TAR. Surgeon’s experience affected the consideration of these clinical factors as contraindications. Conclusions: Total ankle replacement has a substantial and increasing role in the treatment of end-stage ankle arthritis in patients over the age of 60. Absolute and potential contraindications of the procedures were indicated from a cross-sectional survey of AOFAS members. Surgeons more experienced with total ankle replacement felt more comfortable employing it in a wider range of clinical settings. Level of Evidence: Level III, therapeutic.

Download Full-text

Nationwide Analysis of Total Ankle Replacement and Ankle Arthrodesis in Medicare Patients: Trends, Complications, and Cost

Foot & Ankle Specialist ◽

10.1177/1938640020950181 ◽

2020 ◽

pp. 193864002095018

Author(s):

William A. Tucker ◽

Brandon L. Barnds ◽

Brandon L. Morris ◽

Armin Tarakemeh ◽

Scott Mullen ◽

...

Keyword(s):

Surgical Intervention ◽

Ankle Arthrodesis ◽

Total Ankle Replacement ◽

Complication Rates ◽

Ankle Arthritis ◽

Ankle Replacement ◽

Procedure Cost ◽

Medicare Patients ◽

Retrospective Comparative Study ◽

End Stage

Background Surgical management of end-stage ankle arthritis consists of either ankle arthrodesis (AA) or total ankle replacement (TAR). The purpose of this study was to evaluate utilization trends in TAR and AA and compare cost and complications. Methods Medicare patients with the diagnosis of ankle arthritis were reviewed. Patients undergoing surgical intervention were split into AA and TAR groups, which were evaluated for trends as well as postoperative complications, revision rates, and procedure cost. Results A total of 673 789 patients were identified with ankle arthritis. A total of 19 120 patients underwent AA and 9059 underwent TAR. While rates of AA remained relatively constant, even decreasing, with 2080 performed in 2005 and 1823 performed in 2014, TAR rates nearly quadrupled. Average cost associated with TAR was $12559.12 compared with $6962.99 for AA ( P < .001). Overall complication rates were 24.9% in the AA group with a 16.5% revision rate compared with 15.1% and 11.0%, respectively, in the TAR group ( P < .001). Patients younger than 65 years had both higher complication and revision rates. Discussion TAR has become an increasingly popular option for the management of end-stage ankle arthritis. In our study, TAR demonstrated both lower revision and complication rates than AA. However, TAR represents a more expensive treatment option. Levels of Evidence: Level III: Retrospective comparative study

Download Full-text