Orthopedics

Feature Article Supplemental Data

Reasons for Failure of Primary Total Hip Arthroplasty Performed Through a Direct Anterior Approach

Nicholas C. Duethman, MD; Joseph M. Statz, MD; Robert T. Trousdale, MD; Michael J. Taunton, MD

Abstract

The direct anterior approach (DAA) for total hip arthroplasty (THA) is a technique popular among some arthroplasty surgeons. There is currently a paucity of data regarding reasons for failure of THA using the DAA. The authors conducted a retrospective review of prospectively collected data on 56 patients who underwent revision THA at their institution after failing primary THA that was performed through a DAA either at their institution (n=8) or elsewhere (n=48) from January 1, 2010, to June 1, 2017. Patients were grouped by modes of failure and compared using patient characteristics, surgical factors, and radiographic outcomes. Total hip arthroplasties performed through the DAA failed due to infection in 21 (38%) patients, aseptic/mechanical loosening in 14 (25%) patients, intraoperative fracture in 6 (11%) patients, postoperative fracture in 6 (11%) patients, tendinitis or pain of unknown etiology in 3 (5%) patients, metallosis in 3 (5%) patients, instability/dislocation in 2 (4%) patients, and leg length discrepancy in 1 (2%) patient. Patients who underwent revision due to unrecognized intraoperative fracture had a lower body mass index (BMI) and weight than patients who had failure due to postoperative fracture, aseptic loosening, or infection. The 4 most common modes of failure included infection, aseptic loosening, unrecognized intraoperative fracture, and postoperative fracture. Together, these made up 84% of failed DAA THAs. Patients with a lower BMI are more likely to have failure due to intraoperative fractures. Patients with a higher BMI are more likely to have failure due to postoperative fracture, aseptic loosening, or infection. [Orthopedics. 2020;43(4);239–244.]

Abstract

The direct anterior approach (DAA) for total hip arthroplasty (THA) is a technique popular among some arthroplasty surgeons. There is currently a paucity of data regarding reasons for failure of THA using the DAA. The authors conducted a retrospective review of prospectively collected data on 56 patients who underwent revision THA at their institution after failing primary THA that was performed through a DAA either at their institution (n=8) or elsewhere (n=48) from January 1, 2010, to June 1, 2017. Patients were grouped by modes of failure and compared using patient characteristics, surgical factors, and radiographic outcomes. Total hip arthroplasties performed through the DAA failed due to infection in 21 (38%) patients, aseptic/mechanical loosening in 14 (25%) patients, intraoperative fracture in 6 (11%) patients, postoperative fracture in 6 (11%) patients, tendinitis or pain of unknown etiology in 3 (5%) patients, metallosis in 3 (5%) patients, instability/dislocation in 2 (4%) patients, and leg length discrepancy in 1 (2%) patient. Patients who underwent revision due to unrecognized intraoperative fracture had a lower body mass index (BMI) and weight than patients who had failure due to postoperative fracture, aseptic loosening, or infection. The 4 most common modes of failure included infection, aseptic loosening, unrecognized intraoperative fracture, and postoperative fracture. Together, these made up 84% of failed DAA THAs. Patients with a lower BMI are more likely to have failure due to intraoperative fractures. Patients with a higher BMI are more likely to have failure due to postoperative fracture, aseptic loosening, or infection. [Orthopedics. 2020;43(4);239–244.]

The direct anterior approach (DAA) for total hip arthroplasty (THA) is a technique popular among some arthroplasty surgeons. It is designated as a minimally invasive technique and uses a true internervous and intermuscular plane between the sartorius and tensor fascia latae muscles superficially and the rectus femoris and gluteus medius deeply.1,2

In support of the minimally invasive nature of the DAA, multiple studies have found decreased muscle damage and pain using the DAA for THA compared with the posterior approach.3,4 Decreased postoperative dislocation rates have also been reported.5 However, Barret et al6 compared the DAA and the posterior approach and determined that all measurable differences in function equilibrate by 6 months after surgery. Furthermore, additional studies have discussed complications following DAA, including periprosthetic fracture, infections, aseptic loosening, leg length discrepancies, and implant malalignment leading to instability and dislocation.5,7

There is currently a paucity of data in the literature regarding reasons for failure of THA using the DAA. Arthroplasty surgeons would benefit from additional knowledge regarding pitfalls to avoid when performing THA through the DAA and patient risk factors that lead to specific causes of early revision. The purpose of this study was to define the authors' institutional experience with patients undergoing revision THA (RTHA) following DAA for primary THA by characterizing indications for revision surgery and identifying patient factors that correlate with these complications.

Materials and Methods

After receiving institutional review board approval, the authors conducted a retrospective review using prospectively obtained single-institution joint registry data.8 This registry collects clinical and radiographic data postoperatively at 1 year, 2 and 5 years, and every 5 years thereafter. Patients unable to follow up in person are mailed questionnaires with requests to return new radiographs. The authors identified all RTHAs performed between January 1, 2010, and June 1, 2017. They then isolated patients who underwent RTHA who had their primary THA performed using the DAA.

Patient Demographics

The authors identified 2365 RTHAs at their institution between January 1, 2010, and June 1, 2017. Of these 2365 RTHAs, the authors identified 56 RTHAs performed for failure of a primary THA performed through a DAA. They defined revision as return to the operating room for exchange or removal of any implant (liner, acetabular component, head, and/or femoral component). Irrigation and debridements without exchange of components were not included as revision operations. Among these 56 RTHAs, 25 (45%) patients were female, mean body mass index (BMI) was 30.5 kg/m2 (range, 18.9–48.6 kg/m2), and 26 (46%) were right sided. Mean patient age was 63.9 years (range, 20–85 years). Regarding index procedure, 48 (86%) patients had DAA performed at an outside hospital and 8 (14%) had DAA performed at the authors' institution. At the latter between January 1, 2010, and June 1, 2017, primary THA was performed through a DAA on 1299 hips.

Statistical Methods

Data collected in this study were analyzed using JMP software version 10.0.0 (SAS Institute, Inc, Cary, North Carolina). Statistical analysis was conducted using analysis of variance as well as Wilcoxon pairwise comparisons to identify patient characteristics, surgical variables, and radiographic parameters that correlate with certain indications for revision operation. Continuous data were reported as means and ranges.

Radiographic Evaluation

Anteroposterior pelvis, anteroposterior femur, and cross-table lateral radiographs after index procedures were assessed to identify acetabular anteversion, angle of inclination, and femoral alignment. Acetabular anteversion was measured on a cross-table lateral radiograph comparing a vertical line with the profile of the cup. Angle of inclination was measured on anteroposterior pelvis radiographs comparing a line drawn below the ischial tuberosities with the profile of the acetabular component. In cases of revision for aseptic loosening or fracture, radiographs immediately after the index procedure were used for evaluation. An unrecognized intraoperative fracture was diagnosed if postoperative radiographs obtained within 3 months of the procedure showed a peri-prosthetic fracture and the patient had no postoperative traumatic event.

Results

Reasons for Revision

Total hip arthroplasties performed through the DAA failed due to infection in 21 (38%) patients, aseptic/mechanical loosening of the femur in 11 (19.6%) patients (Figure 1), aseptic/mechanical loosening of the acetabulum in 2 (3.6%) patients, aseptic/mechanical loosening of both the femur and the acetabular components in 1 (1.8%) patient, unrecognized intraoperative periprosthetic femoral calcar fracture leading to postoperative stem subsidence in 5 (8.9%) patients (Figure 2), unrecognized intraoperative greater trochanter fracture in 1 (1.8%) patient, postoperative periprosthetic fracture in 6 (10.7%) patients, tendinitis or pain of unknown etiology in 3 (5.4%) patients, metallosis in 3 (5.4%) patients (Figure 3), instability/dislocation in 2 (3.6%) patients (Figure 4), and leg length discrepancy in 1 (1.8%) patient (Figure 5).

Anteroposterior radiograph of the femur obtained after a right total hip arthroplasty through a direct anterior approach demonstrating aseptic loosening of the femoral component.

Figure 1:

Anteroposterior radiograph of the femur obtained after a right total hip arthroplasty through a direct anterior approach demonstrating aseptic loosening of the femoral component.

Anteroposterior radiograph of the pelvis obtained after a left total hip arthroplasty through a direct anterior approach demonstrating a potentially missed intraoperative calcar fracture.

Figure 2:

Anteroposterior radiograph of the pelvis obtained after a left total hip arthroplasty through a direct anterior approach demonstrating a potentially missed intraoperative calcar fracture.

Anteroposterior radiograph of the femur obtained after a left total hip arthroplasty through a direct anterior approach demonstrating a metal-on-metal component.

Figure 3:

Anteroposterior radiograph of the femur obtained after a left total hip arthroplasty through a direct anterior approach demonstrating a metal-on-metal component.

Lateral radiograph of the femur obtained after a right total hip arthroplasty through a direct anterior approach demonstrating anterior instability.

Figure 4:

Lateral radiograph of the femur obtained after a right total hip arthroplasty through a direct anterior approach demonstrating anterior instability.

Pie chart detailing modes of failure in a cohort of patients undergoing revision surgery following total hip arthroplasty through a direct anterior approach.

Figure 5:

Pie chart detailing modes of failure in a cohort of patients undergoing revision surgery following total hip arthroplasty through a direct anterior approach.

Revision Surgeries

Revision THAs following the index DAA were performed through a DAA in 10 (17.9%) patients; an anterolateral approach in 12 (21.4%) patients, with a Wagner osteotomy in 2 (3.6%) of these patients; a transfemoral osteotomy in 9 (16.1%) patients; a posterior approach in 22 (39.3%) patients; and an extended trochanteric osteotomy in 1 (1.8%) patient. Revision procedures consisted of an exchange of head and/or liner in 10 (17.9%) patients, acetabular component revision with or without head exchange in 5 (8.9%) patients, femoral component revision with or without liner exchange in 21 (37.5%) patients, and revision of both femoral and acetabular components in 20 (35.7%) patients.

The authors conducted pairwise comparisons between patients, grouped by failure mode, to elucidate differences in patient demographics. Patient demographics and surgical variables are presented in Table 1.

Summary of Failure Mode Based on Patient Characteristics, Operative Variables, and Radiographic Parameters

Table 1:

Summary of Failure Mode Based on Patient Characteristics, Operative Variables, and Radiographic Parameters

Further details concerning statistically significant pairwise comparisons between groups are presented in Table A (available in the online version of the article). Patients who had failure due to aseptic loosening were taller than those who had failure due to instability (P=.047). Patients who had failure due to postoperative fracture, aseptic loosening, or infection had a higher BMI than those who had failure due to unrecognized intraoperative periprosthetic fracture (P=.0013 to .0306). Time to revision surgery was longer for patients who had failure due to metallosis (5.22 years) and aseptic loosening (2.64 years) as compared with infection (1.14 years; P=.002 to .018). Also, time to revision surgery was longer for patients who had failure due to metallosis (5.22 years), aseptic loosening (2.64 years), and infection (1.14 years) as compared with unrecognized intraoperative fracture (0.24 years; P=.002 to .027). Time to revision surgery was longer for patients who had failure due to aseptic loosening (2.64 years) as compared with postoperative fracture (1.39 years; P=.029).

Significant Pairwise Comparisons by Failure Mode

Table A.

Significant Pairwise Comparisons by Failure Mode

Postrevision Complications

Overall, 11 (19.6%) patients had a complication after RTHA. Of the patients who underwent revision for aseptic loosening (n=14), 2 (14.3%) patients had postrevision complications: 1 had a dislocation and 1 had a pulmonary embolus. Four (19.0%) of 21 patients who underwent revision for infection had postrevision complications, which included wound dehiscence, aseptic loosening, repeat infection, and seroma. One of 3 patients who underwent revision for metallosis had a postrevision hematoma. One patient who underwent revision for pain of unknown etiology had multiple postrevision dislocations. Two (16.7%) of 12 patients who underwent revision for periprosthetic fracture had complications: 1 had a postrevision infection requiring repeat surgery and 1 had stem subsidence that did not require reoperation. One of 2 patients who underwent surgery for tendinitis had a postrevision dislocation. The 2 patients who underwent revision for instability and 1 patient who underwent revision for leg length discrepancy did not have postrevision complications.

Discussion

The DAA may be associated with unique complications during and after THA as compared with THA performed through other approaches. There is currently a paucity of data in the literature regarding reasons for failure of THA using the DAA. The purpose of this study was to define the authors' institutional experience with patients undergoing RTHA following DAA for primary THA by characterizing indications for revision surgery, identifying patient factors that correlate with these complications, and reporting their rate of revision.

Meneghini et al9 recently reported that the DAA increases the risk of early femoral failure. Their study showed that early failure due to femoral fracture or loosening occurred more with the DAA compared with the posterior or direct lateral approaches. They also implicated the DAA and the posterior approach as leading to more early revisions due to instability compared with the direct lateral approach.

The current authors recently compared the DAA with the mini-posterior approach.10 In this study, there was no difference in complication rate in THA completed through the DAA compared with the mini-posterior approach. Patients in the DAA cohort met postoperative milestones slightly sooner, such as walking without a gait aid, discontinuing narcotics, completing stairs, and walking 6 blocks. However, these advantages normalized at 8 weeks.

In the current study, infection, aseptic loosening, unrecognized intraoperative fracture, and postoperative fracture were the 4 most common modes of failure following DAA. When combined, they made up 84% of the indications for revision following this approach. Eto et al11 also reported on the cause of revision following DAA. The most common indication in their cohort was aseptic loosening (36.7%), followed by metallosis (33.4%). Infection was the indication for 10% and fracture was the indication for 10% of their patients. The current authors had a low number of hips revised for metallosis (n=3; 5.4%), which was likely explained by differences in implant materials used during primary procedures. When patients with metallosis are removed from the cohort of Eto et al,11 infection, aseptic loosening, and fracture then make up 85% of their revisions, which aligns with the current study population.

Bhandari et al5 reported their cohort of 1277 DAA hips at maximum follow-up of 3 years. Their overall revision rate was 2.7%, with aseptic loosening (43%), periprosthetic fracture (23.3%), and hip dislocation (20%) accounting for 86.3% of their revisions. Lee and Marconi7 conducted a systematic review of complication rates following THA through the DAA. They reported intraoperative fractures 2.3% of the time, postoperative fractures in 2.8% of cases, and deep infections in 0.6% of cases. With the exception of higher rates of revision for instability reported by Bhandari et al,5 the indication for revision in the current population is similar to that of these 2 studies.

Indications for revision THA following anterolateral, direct lateral, and posterior approaches have been reported.12 Revision for septic indications occurred in 0.5%, 1.1%, and 0.5% of anterolateral, direct lateral, and posterior approach primary cases, respectively, compared with 0.5% of DAA primary cases. Revision rates for aseptic indications were 1.4%, 1.8%, and 1.9% in anterolateral, direct lateral, and posterior THAs, respectively, compared with 1.1% in DAA primary hips.

As expected, unrecognized intraoperative fracture was the indication for earliest revision, at 0.24 years. Logically, patients who have an intraoperative fracture require early revision. For perspective, Matta et al13 reported an intraoperative calcar fracture rate through the DAA of 0.8%. Nakata et al14 reported intraoperative calcar fracture rates of 1.0% in DAA hips and 1.0% in mini-posterior hips. Conversely, complications such as aseptic loosening, infection, metallosis, and postoperative fracture take time to develop and would lead to later revision.

In the current study, a comparison of failure modes revealed interesting differences in patient demographics between these groups. Patients who underwent revision due to unrecognized intraoperative fracture had a lower BMI and weight than patients who had failure due to postoperative fracture, aseptic loosening, or infection (P=.0013 to .0306). This is likely due to several reasons. First, patients who are lighter and thinner likely have more fragile bone and higher rates of osteoporosis, leading to a higher likelihood of intraoperative fracture. Second, patients who are heavier place more force through their implants, which conceptually leads to higher likelihood of postoperative fracture and aseptic loosening. Third, higher BMI is associated with increased risk of infection.15

Additionally, patients who had failure due to aseptic loosening were taller than those who had failure due to instability (P=.047). This could be explained by differences in exposure and component sizes. Taller people have larger femurs, which can make exposure and preparation of the femur and acetabulum more difficult. Because the majority (86%) of the revisions performed at the authors' institution had index DAA THA performed elsewhere, immediate postoperative radiographs following the index procedure were not available for evaluation. This could have led to slight malposition of components, which may lead to a higher rate of aseptic loosening. Regarding instability, shorter individuals have smaller acetabula. Therefore, smaller femoral heads were potentially used in these patients, leading to a smaller head-to-neck ratio and shorter jump distance and thus increased risk of dislocation. However, this is speculative because both patients who underwent revision for instability had primary surgeries at outside facilities, and primary implant sizes are unknown.

This study had the usual limitations associated with a retrospective review. Data were limited regarding index THA, with 48 being performed at outside institutions. The authors do not have the total number of primary DAA THAs performed at outside institutions to calculate the total incidence of revision in their study population. Also, this study did not compare indications for revision surgery when other approaches were used for the index operation, which would have provided valuable information. Not all repeat revisions would come through the authors' institution; therefore, postoperative complication rates could be limited and possibly underestimated.

Conclusion

Overall, this study has brought to light interesting data of use to the arthroplasty surgeon performing THA through the DAA. Care must be taken to avoid the most common complications of infection, aseptic loosening, unrecognized intraoperative fracture, and postoperative fracture. Appropriate counseling of heavier and larger patients regarding their risk of infection, aseptic loosening, and postoperative fracture should be performed preoperatively. Knowing that patients who have failure due to aseptic loosening tend to be taller should encourage surgeons to take extra time to ensure that all components fit appropriately in this population. Maximizing femoral head size and examining stability intraoperatively should be encouraged for shorter patients. Finally, extra care must be taken with lighter and thinner patients to avoid large forces during manipulation and component preparation and insertion that may lead to unrecognized intraoperative fracture.

References

  1. Rachbauer F, Kain MS, Leunig M. The history of the anterior approach to the hip. Orthop Clin North Am. 2009;40(3):311–320. doi:10.1016/j.ocl.2009.02.007 [CrossRef] PMID:19576398
  2. Smith-Petersen MN. A new supra-articular subperiosteal approach to the hip joint. Am J Orthop Surg (Phila Pa). 1917;15:593.
  3. Bergin PF, Doppelt JD, Kephart CJ, et al. Comparison of minimally invasive direct anterior versus posterior total hip arthroplasty based on inflammation and muscle damage markers. J Bone Joint Surg Am. 2011;93(15):1392–1398. doi:10.2106/JBJS.J.00557 [CrossRef] PMID:21915544
  4. Bremer AK, Kalberer F, Pfirrmann CW, Dora C. Soft-tissue changes in hip abductor muscles and tendons after total hip replacement: comparison between the direct anterior and the transgluteal approaches. J Bone Joint Surg Br. 2011;93(7):886–889. doi:10.1302/0301-620X.93B7.25058 [CrossRef] PMID:21705558
  5. Bhandari M, Matta JM, Dodgin D, et al. Anterior Total Hip Arthroplasty Collaborative Investigators. Outcomes following the single-incision anterior approach to total hip arthroplasty: a multicenter observational study. Orthop Clin North Am. 2009;40(3):329–342. doi:10.1016/j.ocl.2009.03.001 [CrossRef] PMID:19576400
  6. Barrett WP, Turner SE, Leopold JP. Prospective randomized study of direct anterior vs postero-lateral approach for total hip arthroplasty. J Arthroplasty. 2013;28(9):1634–1638. doi:10.1016/j.arth.2013.01.034 [CrossRef] PMID:23523485
  7. Lee GC, Marconi D. Complications following direct anterior hip procedures: costs to both patients and surgeons. J Arthroplasty. 2015;30(9)(suppl):98–101. doi:10.1016/j.arth.2015.03.043 [CrossRef] PMID:26118568
  8. Berry DJ, Kessler M, Morrey BF. Maintaining a hip registry for 25 years: Mayo Clinic experience. Clin Orthop Relat Res. 1997;344:61–68. doi:10.1097/00003086-199711000-00007 [CrossRef] PMID:9372758
  9. Meneghini RM, Elston AS, Chen AF, Kheir MM, Fehring TK, Springer BD. Direct anterior approach: risk factor for early femoral failure of cementless total hip arthroplasty. A multicenter study. J Bone Joint Surg Am. 2017;99(2):99–105. doi:10.2106/JBJS.16.00060 [CrossRef] PMID:28099299
  10. Taunton MJ, Trousdale RT, Sierra RJ, Kaufman K, Pagnano MW. John Charnley Award. Randomized clinical trial of direct anterior and miniposterior approach THA: which provides better functional recovery?Clin Orthop Relat Res. 2018;476(2):216–229. doi:10.1007/s11999.0000000000000112 [CrossRef] PMID:29529650
  11. Eto S, Hwang K, Huddleston JI, Amanatullah DF, Maloney WJ, Goodman SB. The direct anterior approach is associated with early revision total hip arthroplasty. J Arthroplasty. 2017;32(3):1001–1005. doi:10.1016/j.arth.2016.09.012 [CrossRef] PMID:27843039
  12. Sheth D, Cafri G, Inacio MC, Paxton EW, Namba RS. Anterior and anterolateral approaches for THA are associated with lower dislocation risk without higher revision risk. Clin Orthop Relat Res. 2015;473(11):3401–3408. doi:10.1007/s11999-015-4230-0 [CrossRef] PMID:25762014
  13. Matta JM, Shahrdar C, Ferguson T. Single-incision anterior approach for total hip arthroplasty on an orthopaedic table. Clin Orthop Relat Res. 2005;441(441):115–124. doi:10.1097/01.blo.0000194309.70518.cb [CrossRef] PMID:16330993
  14. Nakata K, Nishikawa M, Yamamoto K, Hirota S, Yoshikawa H. A clinical comparative study of the direct anterior with mini-posterior approach: two consecutive series. J Arthroplasty. 2009;24(5):698–704. doi:10.1016/j.arth.2008.04.012 [CrossRef] PMID:18555653
  15. Wagner ER, Kamath AF, Fruth KM, Harm-sen WS, Berry DJ. Effect of body mass index on complications and reoperations after total hip arthroplasty. J Bone Joint Surg Am. 2016;98(3):169–179. doi:10.2106/JBJS.O.00430 [CrossRef] PMID:26842406

Summary of Failure Mode Based on Patient Characteristics, Operative Variables, and Radiographic Parameters

Mode of FailureNo. Male/FemaleMean

Age, yHeight, cmWeight, kgBody Mass Index, kg/m2Anesthesia Time, minOperative Time, minTime to Surgery, yAnteversionInclination
Aseptic loosening4/966.1174.097.431.8239.2157.42.630.6°46.4°
Infection11/1064.3171.293.131.6198.4129.21.129.1°43.3°
Instability2/040.0154.566.528.1200.0117.00.638.0°36.2°
Intraoperative fixation5/171.2169.264.822.5193.5127.70.228.3°40.2°
Postoperative fracture2/467.8170.093.732.1193.8124.01.423.5°44.0°
Metallosis1/248.3180.793.728.4195.0118.05.235.2°44.8
Tendonitis0/261.0184.0114.533.2160.594.06.620.8°48.3°
Pain of unknown etiology0/160.0184.0107.031.6197.0141.05.027.0°41.3°
Leg length discrepancy0/161.0175.095.031.0198.0144.01.526.0°44.5°
P.1400.0198.1845.1008.1488.7433.7230.0002.5540.4477

Significant Pairwise Comparisons by Failure Mode

Height (cm)P - value
Aseptic looseningInstability
174.00154.000.047
Weight (kg)P - value
Post-operative fractureIntra-operative fracture
93.7064.800.031
Aseptic looseningIntra-operative fracture
97.4064.800.011
InfectionIntra-operative fracture
93.1064.800.004
BMI (kg/m2)P - value
Post-operative fractureIntra-operative fracture
32.1022.400.013
Aseptic looseningIntra-operative fracture
31.8022.500.001
InfectionIntra-operative fracture
31.6022.500.003
Time to Surgery (yrs)P - value
MetallosisInfection
5.221.140.018
Aseptic looseningInfection
2.641.140.002
MetallosisIntra-operative fracture
5.220.240.027
Aseptic looseningIntra-operative fracture
2.640.240.002
InfectionIntra-operative fracture
1.140.240.009
Aseptic looseningPost-operative fracture
2.641.390.029
Authors

The authors are from the Department of Orthopedic Surgery, Mayo Clinic, Rochester, Minnesota.

Drs Duethman, Statz, and Trousdale have no relevant financial relationships to disclose. Dr Taunton is a paid consultant for and receives royalties from DJO Surgical.

Correspondence should be addressed to: Michael J. Taunton, MD, Department of Orthopedic Surgery, Mayo Clinic, 200 First St SW, Rochester, MN 55905 ( Taunton.michael@mayo.edu).

Received: February 08, 2019
Accepted: May 14, 2019
Posted Online: June 05, 2020

10.3928/01477447-20200521-01

Sign up to receive

Journal E-contents