Accepted Manuscript. To appear in: The Journal of Arthroplasty. Received Date: 6 July 2018 Revised Date: 22 August 2018 Accepted Date: 24 August 2018

Transcription

1 Accepted Manuscript The Utilization of Metal Augments Allows Better Biomechanical Reconstruction of the Hip in Revision Total Hip Arthroplasty With Severe Acetabular Defects: A Comparative Study Baochun Zhou, MD, Yixin Zhou, MD, PhD, Dejin Yang, MD, Hao Tang, MD, MRCS, Hongyi Shao, MD, Yong Huang, MD PII: DOI: S (18) /j.arth Reference: YARTH To appear in: The Journal of Arthroplasty Received Date: 6 July 2018 Revised Date: 22 August 2018 Accepted Date: 24 August 2018 Please cite this article as: Zhou B, Zhou Y, Yang D, Tang H, Shao H, Huang Y, The Utilization of Metal Augments Allows Better Biomechanical Reconstruction of the Hip in Revision Total Hip Arthroplasty With Severe Acetabular Defects: A Comparative Study, The Journal of Arthroplasty (2018), doi: / j.arth This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

2 The Utilization of Metal Augments Allows Better Biomechanical Reconstruction of the Hip in Revision Total Hip Arthroplasty With Severe Acetabular Defects: A Comparative Study Baochun Zhou, MD a, Yixin Zhou, MD, PhD a, Dejin Yang, MD a, Hao Tang, MD, MRCS a, Hongyi Shao, MD a, Yong Huang, MD a a. Department of Orthopaedic Surgery Beijing Jishuitan Hospital, Fourth Clinical College of Peking University No.31 Xinjiekou East Street, Xicheng District, Beijing China Please address all correspondence to: Yixin Zhou, MD, PhD Department of Orthopaedic Surgery Beijing Jishuitan Hospital, Fourth Clinical College of Peking University No.31 Xinjiekou East Street, Xicheng District, Beijing China Phone: , FAX: orthoyixin@yahoo.com s of other authors: Baochun Zhou, MD: orthobaochun@163.com Dejin Yang, MD: fafnirdj@163.com Hao Tang, MD, MRCS: pkuhao@163.com Hongyi Shao, MD: jsszshy@163.com Yong Huang, MD: pku @126.com

3 1 2 The Utilization of Metal Augments Allows Better Biomechanical Reconstruction of the Hip in Revision 3 4 Total Hip Arthroplasty With Severe Acetabular Defects: A Comparative Study 1

4 5 Abstract 6 7 Background: Reconstructing the normal hip biomechanics is important for a successful revision total hip arthroplasty (THA). Little is known about whether using metal augments in revision THA is biomechanically superior to traditional techniques. Methods: A retrospective review was conducted on 74 consecutive THAs revised using metal augments with a cementless hemispherical cup and 77 consecutive THAs revised using the jumbo cup, all with a minimum 2-year follow-up. Biomechanical parameters were measured before and immediately after the revision. Radiological and clinical outcomes at follow-ups were also evaluated. Results: The metal augment group had a reconstructed center of rotation (COR) that was 6.5 mm closer to the anatomic COR in height (p<0.001), 3.6 mm smaller cup size (p<0.001), and 5.7 mm less head-cup difference (p<0.001). Moreover, there was a reconstructed COR that was much closer to the anatomic COR (vertical distance, 1.8 mm vs 14.1 mm, p<0.001; horizontal distance, -2.1 mm vs 7.9 mm, p=0.013), 4.1 mm greater femoral offset (p=0.006), and 8 mm less LLD (p=0.035) in the subgroup of Paprosky type III bone defects when compared to the jumbo cup group. All cup-augment constructs were radiologically stable with a higher mean postoperative Harris Hip Score (p=0.012). One jumbo cup was radiologically unstable. Conclusion: In revision THA, utilizing metal augments helps to restore the COR position more precisely, avoid using a larger cup, reduce head-cup difference, rebuild femoral offset, and decrease LLD, particularly with Paprosky type III bone defects Moreover, it provides satisfactory radiological and clinical outcomes in the short-term. 2

5 29 30 Keywords: revision total hip arthroplasty; biomechanical reconstruction; augment; jumbo cup; center of rotation; head-cup difference. 31 3

6 One critical goal of revision total hip arthroplasty (THA) is to attain optimal biomechanical reconstruction of the hip. Hip biomechanics influence abductor muscle function, soft tissue balancing, joint reactive forces, range of motion, liner wear rate, implant stability, gait and, consequently, patient satisfaction and clinical outcomes [1-7]. Radiological evaluation such as the position of the center of rotation (COR), as well as cup orientation, femoral offset, and leg length discrepancy (LLD) is an established method for assessing the biomechanical reconstruction of the hip [8-10]. Even small changes in these parameters can affect clinical outcomes considerably. In this regard, one previous study evaluated the cup durability following primary THA (107 uncemented cups, 55 cemented cups) with acetabular protrusion and found that the rate of aseptic cup revision increases by 24% (hazard ratio, 1.24) if the COR deviates from the anatomic position by 1 mm laterally [2]. Another study on primary THA (145 cemented cups) with congenital dysplasia of the hip demonstrated that the rates of cup loosening and revision increase significantly if this deviation reaches 15 mm of elevation [11]. However, severe acetabular bone loss is frequently encountered during revision THA and jeopardizes optimal biomechanical reconstruction. With substantial bone loss, a standard-size hemispherical cup implanted in the anatomic position is unable to achieve mechanical stability even with multiple screws [12]. In order to obtain initial stability and sufficient cup-bone contact, surgeons may need to apply diverse techniques, including the use of jumbo cups [13], structural allografts [14], impaction bone grafting [15], and cup-cage constructs [16], but still at the risk of compromising biomechanical reconstruction. Recently, jumbo cups and emerging metal augments are the two most popular options for revision THA. The jumbo cup technique, as a classic revision strategy, has been widely used for 4

7 decades owing to its straightforward technique and encouraging outcomes in selected hip revision cases [13, 17, 18]. The concept of this technique was proposed as placement of an extra-large hemispherical acetabular component with an extensive porous coating to engage adequate host bone and achieve both primary mechanical stability and long-term biological fixation [19]. The principle of the jumbo cup is to obtain either rim-fixation or, in most cases, three-point fixation with the host bone [13]. In order to ensure good bone purchase, however, the jumbo cup technique requires that surgeons ream up to a significantly larger cup size, thus removing more host bone [13], and may cause a change in the COR position. Metal augments, due to the modular nature and versatility in shape and size, allow surgeons to fill various bony defects and also enable them to restore the COR position [20, 21]. Consequently, metal augments possess potential advantages in biomechanical reconstruction. In fact, the utilization of metal augments facilitates the rebuilding of the acetabular ring or reconstructing mechanically supportive points without excessively reaming residual host bone. While using metal augments with a cementless hemispherical cup has achieved promising short- to mid-term results [20, 22, 23], few previous studies have verified its superiority over other traditional techniques in biomechanical reconstruction of the hip. The main objective of this study was to evaluate the ability of the metal augment technique to reconstruct normal biomechanics of the hip in revision THA by comparing with traditional jumbo cup technique. The radiographic parameters evaluated consisted of (1) vertical and horizontal positions of the COR; (2) cup size and head-cup difference (size difference between head and cup); (3) femoral offset; (4) LLD; and (5) cup orientation. The secondary objective was to evaluate the radiological and clinical outcomes of these two reconstructive techniques. We 5

8 hypothesized that using metal augments with a hemispherical cementless cup would allow better biomechanical reconstruction of the hip in revision THA and provide satisfactory radiological as well as clinical results when compared to the jumbo cup technique. Patients and Methods Patient Selection This retrospective comparative study received the approval of Institutional Review Board. We consecutively included 76 revision THAs using metal augments with a cementless hemispherical cup between April 2014 and August 2016, and 80 consecutive cementless revision THAs using the jumbo cup technique performed between January 2010 and August 2016 for comparison, all from a single institution. A jumbo cup is recognized as one that is at least 10 mm larger than the cup employed in primary THA [19]. Therefore, we defined the minimum size of a jumbo cup as 64 mm in men and 60 mm in women by adding 10 mm to the mean cup size of all concurrent primary THAs in the same institutional database. This definition was consistent with another Asian population study published previously [24]. Two patients in the metal augment group and three patients in the jumbo cup group failed to complete the minimum 2-year follow-up, but there was no evidence of acetabular failure according to their last follow-up data. This left 74 patients and patients with corresponding mean follow-up time points of 35 months (range, months) and 52 months (range, months) respectively

9 107 Surgical Technique All revisions were conducted through the posterolateral approach. The acetabulum was reamed sequentially into an approximate hemispherical shape until reaching viable host bone. For jumbo cup cases, a substantial amount of host bone was removed to pursue an adequate contact area of supportive bone. Based on the location and extent of defects, it may require reaming at non-anatomic positions. The jumbo cup was secured with rim fixation or three-point fixation. For severe superior or medial bone loss (Paprosky type III), bulk allograft (2 hips) was used to provide structural support for the jumbo cups and impaction bone grafting (12 hips) was used to fill residual bone defects. For metal augment cases, the acetabulum was reamed one or two sizes larger at the anatomic level, with the final reamer positioned as a cup trial. The bone defect was then reassessed, and the type and size of augments were determined according to the shape and extent of the defects around the trial. Appropriate metal augments (Trabecular Metal; Zimmer Inc, Warsaw, IN, USA or AK Medical Limited, Beijing, China) were then used to restore the acetabular rim or reconstruct mechanically supportive points. In cases with severe superior bone loss, extended fixation to the ilium was achieved with either a buttress augment (9 hips) or two stacked slope augments (2 hips). For severe bone loss at the ischium or pubis, considering the shape and size of traditional augments did not fit the bone s geometry, extended fixation was obtained using either trimmed traditional augments (17 hips) (Fig. 1A-D) or a new-type lotus augment (AK Medical Limited, Beijing, China) (2 hips) (Fig. 1E-F, and Fig. 2A-B). After the augments were inserted, a cementless hemispherical cup was implanted and fixed with multiple screws that had adequate purchase into bone. 7

10 Bone cement was used in the cup-augments interface before final cup implantation. Simultaneous revision of the femoral stem was also performed, if indicated. When the stem was preserved, only the femoral head was exchanged. An appropriate head length was chosen to optimize the joint stability and soft tissue tension. The femoral heads used in the metal augment group were larger (33.8 mm vs 31.7 mm; p<0.001) than in the jumbo cup group (Table 1). Biomechanical Evaluation Radiographic Preparation Both preoperative and immediate postoperative anteroposterior radiographs of bilateral hips were reviewed. Well-trained and experienced technicians obtained these radiographs in a standardized manner. The X-ray projection was centered over the patients symphysis pubis with a beam-patient distance of 100 cm. Two authors (name not provided for blinded review) categorized the acetabular bone loss independently according to the Paprosky classification system [25]. In case of any disagreement, a senior author (name not provided for blinded review) was consulted. One author (name not provided for blinded review) who did not participate in the surgeries made all measurements using Mimics 16.0 software (Materialise, Leuven, Belgium). Intraobserver reliability of radiographic measurements indicated strong agreement (Intraclass Correlation Coefficient = 0.894). The magnifications were calculated based on the known head sizes We connected the lowest points of the bilateral teardrops as the horizontal reference line. If the teardrop on the revised side was violated, the contralateral one was mirrored instead. The vertical reference line was perpendicular to the horizontal 8

11 reference line and passed through the lowest point of the ipsilateral teardrop. Both the preoperative COR and reconstructed COR were determined using a best-fit circle aligned with the femoral head margin. To evaluate the precision of COR reconstruction, we mirrored the anatomic COR of the normal contralateral hip to the revised side as a reference landmark. If the contralateral hip was abnormal, we instead used the Ranawat triangle method to determine the anatomic COR, which is based on the theory that the height ratio of the anatomic acetabulum to pelvis is approximated as 20% [26]. Radiographic Measurements We measured the positions of the reconstructed COR referring to the ipsilateral teardrop and anatomic COR as the absolute and relative positions, respectively. The absolute positions were determined as the vertical and horizontal distances from the reconstructed COR to the horizontal and vertical reference lines. The relative positions were determined as the vertical and horizontal distances from the reconstructed COR to the anatomic COR (Fig. 3). Cup size and head-cup difference were determined using known component sizes. Femoral offset was measured as the distance from the reconstructed COR to the anatomic axis of the proximal femoral shaft (Fig. 3). To assess LLD, we measured the distance from the tip of lesser trochanter to the horizontal reference line bilaterally (Fig. 3). A positive value indicated a limb length longer on the revised side than the contralateral side, and vice versa. To ascertain the impact of adjustment of the femoral side on limb length, we measured the vertical distance from the COR to a line passing through the tip of ipsilateral lesser trochanter and parallel to the horizontal reference line on the revised side. Then, we calculated the change of this distance before and after revision; a positive value indicated a limb length elongation adjusted by the 9

12 femoral side, and vice versa [27]. We also measured cup inclination and anteversion angles. Cup anteversion angle was calculated as the result of Arcsin (short axis/long axis), as described by Widmer et al [28] Ultimately, the following radiographic parameters associated with biomechanical reconstruction were analyzed: (1) the vertical and horizontal positions of the COR; (2) cup size and head-cup difference; (3) femoral offset; (4) LLD; (5) cup inclination and anteversion angles. Radiological and Clinical Follow-up Patients were scheduled to revisit the institution at 3, 6 and 12 months, and then yearly after the revision surgery. Patients who did not return were interviewed by telephone and invited to send their follow-up radiographs obtained outside of our institution. Anteroposterior and lateral radiographs of the hip obtained at each follow-up visit were collected to evaluate the radiological stability of acetabular components. Radiolucent lines around the cup and augments were measured and located based on the method proposed by Delee et al [29]. The stability of acetabular construct was evaluated and classified as unstable, or stable with fibrous ingrowth, or stable with bone ingrowth [21, 30]. Radiological failure was defined if the acetabular construct was evaluated as unstable. The Harris Hip Score (HHS) at the last follow-up was used to assess the hip function [31]. Postoperative complications and reoperations were also recorded at each follow-up. Clinical failure was defined as the demand for further acetabular revision for any reason. 10

13 Statistical Analysis Statistical analyses were conducted with the use of SPSS software package (version 15.0, IBM Corp., Armonk, New York, USA), with a p-value <0.05 indicating statistical significance. Categorical variables were summarized as incidences with percentages, and continuous variables were expressed as means with standard deviations, except as otherwise noted. Categorical variables were compared using Chi-square test or Fisher s exact test. Continuous variables were compared via paired t-test or Wilcoxon s signed ranks test, or two group independent t-test or Mann-Whitney U test. Variances of the relative positions of the reconstructed COR were compared using Levene s test. Correlation between variables was evaluated with the Pearson correlation coefficient. Implant survival was assessed using Kaplan-Meier survival analysis with 95% confidence intervals (CI). Prior to the study, a sample size calculation was performed with α=0.05 and power=90%. According to previous case series [20, 23, 32], thirty-two patients in each group were sufficient to detect statistically significant difference in the relative vertical position of the reconstructed COR. Results The two patient groups shared comparable demographic profiles, with the exception that the metal augment group had lower preoperative HHS (39.3 vs 46.7; p<0.001) and more Paprosky type III defects (73%). By comparison, the jumbo cup group had more Paprosky type II defects (71.5%; p<0.001; Table 1 and Table 2). 11

14 Vertical and Horizontal Positions of the COR Vertical Position of the COR The reconstructed COR in the metal augment group was located not only inferiorly (18.9 mm vs 24.1 mm; p<0.001) but also closer to the anatomic COR (3.1 mm vs 9.6 mm; p<0.001) as compared to the jumbo cup group (Table 3). However, the preoperative COR was located superiorly (p=0.006) and further away from the anatomic COR (p=0.033) in the metal augment group (Table 3). Further, we stratified the hips using the Paprosky classification. In Paprosky type III cases, the difference in vertical position of the COR between the two groups was an average of 12.3 mm (p<0.001; Fig. 4A) relative to the anatomic COR. In addition, there was less variation in the relative vertical position in the metal augment group (p=0.003; Fig. 4A). By comparison, in Paprosky type II cases, this difference was 4.1 mm on average (p=0.015; Table 4; Fig. 5A). Horizontal Position of the COR Overall, there was no difference in the horizontal positions of the reconstructed COR between the two groups (p=0.915 and p=0.510, respectively; Table 3). However, in Paprosky type III cases, the reconstructed COR in the metal augment group was located less lateral (30.7 mm vs 35.1mm; p=0.006) and closer to the anatomic COR (-2.1 mm vs 7.9 mm; p=0.013; Fig. 4B) when compared with the jumbo cup group (Table 4); The variation of relative horizontal positions in the metal augment group was also less (p=0.014; Fig. 4B). There was a moderate positive correlation between the vertical and horizontal positions of the reconstructed COR (Pearson correlation coefficient, r=0.444, p<0.001). 12

15 Cup Size and Head-cup Difference The cup size in the metal augment group was smaller (58.2 mm vs 61.8 mm; p<0.001), in addition to a smaller head-cup difference (24.4 mm vs 30.1 mm; p<0.001) when compared to the jumbo cup group (Table 3). The differences were similar after stratification via the Paprosky classification (Table 4). Femoral Offset On the whole, there was no difference in postoperative femoral offset between the two groups (p=0.292) (Table 3). However, in Paprosky type III cases, the postoperative femoral offset was larger in the metal augment group (34.5 mm vs 30.3 mm; p=0.006) by comparison to the jumbo cup group (Table 4). LLD Overall, there was no difference in postoperative LLD between the two groups (p=0.239) (Table 3). The limb length adjustment on the femoral side was larger in the jumbo cup group than in the metal augment group (6.3 ± 14.1 mm vs -0.7 ± 14.2 mm; p=0.003). In Paprosky type III cases, the postoperative LLD was less in the metal augment group (-3.8 mm vs mm; p=0.035) by comparison to the jumbo cup group (Table 4). There was a moderate negative correlation between LLD and the vertical position of COR (Pearson correlation coefficient, r=-0.534, p<0.001)

16 282 Cup Inclination and Anteversion Angles The cup inclination and anteversion angles didn t differ between the two groups (Tables 3 and 4). Radiological and Clinical Outcomes All cup-augment constructs demonstrated bone ingrowth and were considered stable. However, one jumbo cup (1.3%) was unstable at the 30-month follow-up, and additional two (2.6%) were stable with fibrous ingrowth. At the last follow-up, the metal augment group demonstrated a higher HHS than the jumbo cup group (86.7 ± 7.4, range , interquartile vs 83.1 ± 9.0, range 49-97, interquartile 80-90; p=0.012). No differences between groups were detected in the rates of either postoperative complications or reoperations (Table 5). With radiological or clinical failure as the endpoint, the 4-year cumulative survivorship did not differ between the metal augment group (98.5%; 95% CI, 89.6%-99.8%) and the jumbo cup group (94.2%; 95% CI, 81.9%-98.2%; p=0.539; Fig. 6). Discussion Optimal biomechanical reconstruction of the hip is critical to achieving successful outcomes after revision THA [1-7]. A growing body of evidence is proving the efficacy of utilizing metal augments in revision THA with excellent clinical and radiological results [20, 22, 23, 30, 33]. However, there is paucity of information regarding the biomechanical advantages of using metal augments in revision THA. 14

17 Therefore, we sought to compare the biomechanical reconstruction of utilizing metal augments with a cementless hemispherical cup to a traditional jumbo cup technique and evaluate their radiological and clinical outcomes at a minimum 2-year follow-up. The most important finding of this study is that revision with metal augments provides a more anatomical position of COR, smaller cup size, decreased head-cup difference, improved femoral offset and less LLD, particularly for Paprosky type III defects. Vertical and Horizontal Positions of the COR This study revealed that the metal augment technique effectively reduces the COR to an anatomic level. This effect is more significant with the increased severity of bone loss. In Paprosky type III cases, the average vertical position of the reconstructed COR was only 1.8 mm above the anatomic COR, while in the jumbo cup group, this position was 14.1 mm, which was similar to the results reported in previous case series [22, 23, 32, 34-36]. With the metal augment technique, metal augments were utilized to reconstitute the acetabular rim or reconstruct mechanically supportive points. This allowed extended fixation to good bone stock remote to the original acetabulum, enabling us to prioritize the COR during cup reconstruction. While with traditional jumbo cup technique, larger cups were required to engage distant host bone for reliable support. In many cases, although the cup is large enough to fill the acetabulum, the cup must be positioned superiorly or inferiorly to achieve secure bony fixation. Even if the jumbo cup can be placed along the inferior acetabulum intentionally, the COR is often higher than the anatomic hip center due to the increased geometric radius of the jumbo cup [36, 37]. In addition to avoiding a higher COR, we also found that the metal augment 15

18 technique helped to prevent a lower COR (Fig. 5A). This is a result of various metal augments being used at the level of the ischium and/or pubis to achieve primary mechanical stability and long-term biological fixation at the inferior segment of acetabulum (19 hips, 25.7%). The metal augment technique in the current study also offered improved horizontal positioning of the COR. With the metal augment technique, we optimized the COR horizontally via medial wall reconstruction with one restrictor augment or two combined slope augments. For Paprosky type III cases, the reconstructed COR was significantly lateralized in the jumbo cup group (average 7.9 mm lateral to anatomic COR) compared to the augment group (average 2.1 mm medial to anatomic COR). The lateralized COR of the jumbo cups was also associated with the superior position of the cups (Pearson correlation coefficient, r=0.444, p<0.001). Restoring the COR to an anatomic position is imperative to hip biomechanics and component stability. In a previous three-dimensional biomechanical model, Delp et al [38] observed that two cm superior and two cm lateral migration of the COR from an anatomic position reduces the abductor muscle lever arm by an average of 28%. The horizontal position of the COR determines the lever arm of both body weight and the abductors, as well as controls the joint reactive force, thus affecting the abductor function and bearing surface wear [4]. In a clinical study of primary THA with acetabular protrusion, Baghdadi et al [2] documented that the risk of aseptic cup failure increases by 24% when the COR is lateralized by 1 mm from the anatomic position. In the present study, Paprosky type III cases that utilized the jumbo cup technique led to an average of 12.3 mm in elevation and 10 mm of lateralization in the COR compared to the metal augment technique. According to the mathematic model deduced by Goran et al [39], this resultant relative migration increased hip loads by 16

19 %. Furthermore, the COR in Paprosky type III cases reconstructed with the jumbo cup technique was approximately 15 mm superior to the anatomic COR, which was a predictor of component loosening and revision on both the acetabular and femoral sides [11]. Cup size and Head-cup Difference Using metal augments assists with avoiding the use of larger cup sizes and thereby decreasing the head-cup difference. Cup size and head-cup differences influence hip stability after THA [40, 41]. Hip dislocation is one of the major concerns of using the jumbo cup technique [42]. Oversized cups also potentially result in iliopsoas impingement and groin pain after THA [43]. The traditional jumbo cup technique requires additional bone reaming to obtain adequate healthy host bone and requires the use of larger cup sizes [19]. In contrast, using metal augments requires less host bone to be removed and fills defects more effectively, so that larger cup sizes are no longer needed given that the volume of bone loss is less [44]. Reduced head-cup differences decrease the space between the femoral head and the pseudo capsule, along with other periarticular soft tissues, which facilities the surrounding soft tissues to maintain hip joint stability. Metal augments do enlarge joint space as well, but the enlarged space is focal and limited, while a joint with a larger head-cup difference has a universally larger joint space. 379 Femoral Offset For Paprosky type III cases, reasonable femoral offset was enabled because the 17

20 reconstructed COR approached the anatomic COR (average 1.8 mm vertically and -2.1 mm horizontally) if metal augments were properly used. However, femoral offset had to be compromised in some cases since the COR required significant lateralization (7.9 mm on average) if the jumbo cup technique was used for cup revision. LLD Using metal augments assists with controlling LLD for patients with substantial acetabular bone loss. For Paprosky type III cases included in the present study, the metal augment group had a mean LLD of 3.8 mm after revision, as compared to more than 1 cm LLD in the jumbo cup group, which is large enough to cause gait deviation [45]. The reason to this difference lies in the distinct vertical position of COR: Using metal augments reduces the COR to a near-normal height, hence much less compensation (average 0.7mm shortening) is required on the femoral side, in comparison with that of jumbo cup group (average 6.3 mm elongation required in the femoral side) since the COR reduction is inadequate as compared to the severe preoperative LLD. Although the limb length can be further lengthened by adjusting femoral side length, both techniques resulted in mild residual limb length shortening, which can be attributed to the following possible reasons: (1) most revision patients have soft tissue contracture due to surgical scars limiting the limb length restoration [27]; (2) evaluating limb length intraoperatively is usually challenging given pelvis deformities and the lack of reliable bony landmarks [27]. 18

21 Cup Inclination and Anteversion Angles Cup abduction and anteversion angles were similar between groups, and these average angles fell within the safe zone for both groups [46]. Previous studies reported similar results of cup abduction angles (42-45 ) for the metal augment technique [21, 23, 36] and the jumbo cup technique (43-48 ) [24, 34, 47, 48]). Based on the above evidence, it is reasonable to infer that both techniques could offer optimal cup orientation in appropriate cases of revision THAs. Radiological and Clinical Outcomes The biomechanical superiorities associated with the metal augment technique were reflected in the radiological and clinical outcomes. While all cup-augment constructs were radiologically stable, one jumbo cup developed radiological failure. Clinically, there were 4 dislocations and 1 reoperation for aseptic loosening in the jumbo cup cases. The failure and complications may occur due to the deviated CORs [2, 11], oversized cups [40, 41], and other poor biomechanics associated with jumbo cups. Concerning hip function, the patients receiving metal augments demonstrated a higher HHS at the last follow-up, though their preoperative level was lower as compared to the jumbo cup group. Therefore, the patients receiving metal augments achieved more improvement in HHS, which may result from the improvements in hip biomechanics. Nevertheless, the mean postoperative HHS in two groups were similarly good (fell within 80-90) [49], which indicated that both techniques yielded satisfactory and comparable hip function. 19

22 Although the metal augment technique provides encouraging short- to mid-term outcomes based on the current study and previous reports [20-23, 30, 33-35], it is important to note that the augment-cement-cup interface may generate debris over time, which may cause component failure and systematic adverse reactions in the long-term [34, 50, 51]. Besides, the jumbo cup technique produces acceptable biomechanical results and promising outcomes in cases with Paprosky type II bone defects, thereby continuing to be an option in the mild defect setting. Limitations of the Study The current study had several limitations which should be emphasized. First, due to the nonrandomized nature, we had more Paprosky type III cases in the metal augment group than the jumbo cup group. Nonetheless, the data in the current study demonstrated that metal augments, combined with a cementless hemispherical cup, provides biomechanical parameters similar or superior to the jumbo cup technique, even with more severe bone loss. This suggests that metal augments in the setting of a cementless hemispherical cup may possess an advantage over the jumbo cup technique with respect to biomechanical reconstruction of the hip especially in Paprosky type III cases with extensive acetabular bone defects. Second, this is a retrospective study. Due to the availability of large femoral head, the head-cup difference in the jumbo cup group can be artificially enlarged by using a relatively small femoral head. However, we utilized the outer diameter (OD) of the hemispherical metal shell in our calculations, and the cup OD was significantly smaller in metal augment group than the jumbo cup group, which reduced the head-cup difference. 20

23 Conclusion In summary, utilizing metal augments with a cementless hemispherical cup in revision THA allows better biomechanical reconstruction of the hip, particularly in the presence of Paprosky type III bone defects. Superiorities in biomechanical reconstruction include a more anatomically positioned COR, less head-cup difference, improved femoral offset, and less LLD, which are reflected in the satisfactory radiological and clinical outcomes in the short-term. Long-term follow-up results are warranted to further determine whether patients will benefit from the improved biomechanics associated with the metal augment technique. 21

24 References [1] Asayama I, Chamnongkich S, Simpson KJ, Kinsey TL, Mahoney OM. Reconstructed hip joint position and abductor muscle strength after total hip arthroplasty. J Arthroplasty 2005; 20(4): [2] Baghdadi YM, Larson AN, Sierra RJ. Restoration of the hip center during THA performed for protrusio acetabuli is associated with better implant survival. Clin Orthop Relat Res 2013; 471(10): [3] Charles MN, Bourne RB, Davey JR, Greenwald AS, Morrey BF, Rorabeck CH. Soft-tissue balancing of the hip: the role of femoral offset restoration. J Bone Joint Surg Am 2004; 86(5): [4] Johnston RC, Brand R, Crowninshield R. Reconstruction of the hip. A mathematical approach to determine optimum geometric relationships. J Bone Joint Surg Am 1979; 61(5): [5] Little NJ, Busch CA, Gallagher JA, Rorabeck CH, Bourne RB. Acetabular polyethylene wear and acetabular inclination and femoral offset. Clin Orthop Relat Res 2009; 467(11): [6] Mcgrory BJ, Morrey BF, Cahalan TD, An K-N, Cabanela M. Effect of femoral offset on range of motion and abductor muscle strength after total hip arthroplasty. J Bone Joint Surg Br 1995; 77(6): [7] Whitehouse MR, Stefanovich-Lawbuary NS, Brunton LR, Blom AW. The impact of leg length discrepancy on patient satisfaction and functional outcome following total hip arthroplasty. J Arthroplasty 2013; 28(8): [8] Silva M, Lee KH, Heisel C, dela Rosa MA, Schmalzried TP. The biomechanical results of total hip resurfacing arthroplasty. J Bone Joint Surg Am 2004; 86(1): [9] Girard J, Lavigne M, Vendittoli P-A, Roy A. Biomechanical reconstruction of the hip: a randomised study comparing total hip resurfacing and total hip arthroplasty. J Bone Joint Surg Br 2006; 88(6): [10] Schmidutz F, Beirer M, Weber P, Mazoochian F, Fottner A, Jansson V. Biomechanical reconstruction of the hip: comparison between modular short-stem hip arthroplasty and conventional total hip arthroplasty. Int Orthop 2012; 36(7): [11] Pagnano MW, Hanssen AD, Lewallen DG, Shaughnessy WJ. The effect of superior placement of the acetabular component on the rate of loosening after total hip arthroplasty. Long-term results in patients who have Crowe Type-II congenital dysplasia of the hip. J Bone Joint Surg Am 1996; 78(7): [12] Sporer SM, Paprosky WG, O'rourke M. Managing bone loss in acetabular revision. J Bone Joint Surg Am 2005; 87(7): [13] Gustke KA, Levering MF, Miranda MA. Use of jumbo cups for revision of acetabulae with large bony defects. J Arthroplasty 2014; 29(1): [14] Brown NM, Morrison J, Sporer SM, Paprosky WG. The use of structural distal femoral allograft for acetabular reconstruction of Paprosky type IIIA defects at a mean 21 years of follow-up. J Arthroplasty 2016; 31(3): [15] Schmitz MW, Hannink G, Gardeniers JW, Verdonschot N, Slooff TJ, Schreurs BW. Acetabular Reconstructions with Impaction Bone-Grafting and a Cemented Cup in Patients Younger Than 50 Years of Age: A Concise Follow-up, at 27 to 35 Years, of a Previous Report. J Bone Joint Surg Am 2017; 99(19): [16] Sculco PK, Ledford CK, Hanssen AD, Abdel MP, Lewallen DG. The Evolution of the Cup-Cage Technique for Major Acetabular Defects: Full and Half Cup-Cage Reconstruction. J Bone Joint Surg Am 2017; 99(13): [17] von Roth P, Abdel MP, Harmsen WS, Berry DJ. Uncemented jumbo cups for revision

25 total hip arthroplasty: a concise follow-up, at a mean of twenty years, of a previous report. J Bone Joint Surg Am 2015; 97(4): [18] Wedemeyer C, Neuerburg C, Heep H, Von Knoch F, Von Knoch M, Löer F, Saxler G. Jumbo cups for revision of acetabular defects after total hip arthroplasty: a retrospective review of a case series. Arch Orthop Trauma Surg 2008; 128(6): [19] Whaley AL, Berry DJ, Harmsen WS. Extra-large uncemented hemispherical acetabular components for revision total hip arthroplasty. J Bone Joint Surg Am 2001; 83(9): [20] Whitehouse MR, Masri BA, Duncan CP, Garbuz DS. Continued good results with modular trabecular metal augments for acetabular defects in hip arthroplasty at 7 to 11 years. Clin Orthop Relat Res 2015; 473(2): [21] Weeden SH, Schmidt RH. The use of tantalum porous metal implants for Paprosky 3A and 3B defects. J Arthroplasty 2007; 22(6): [22] O'Neill CJ, Creedon SB, Brennan SA, O'Mahony FJ, Lynham RS, Guerin S, Gul R, Harty JA. Acetabular Revision Using Trabecular Metal Augments for Paprosky Type 3 Defects. J Arthroplasty 2018; 33(3): [23] Jenkins DR, Odland AN, Sierra RJ, Hanssen AD, Lewallen DG. Minimum Five-Year Outcomes with Porous Tantalum Acetabular Cup and Augment Construct in Complex Revision Total Hip Arthroplasty. J Bone Joint Surg Am 2017; 99(10): e49. [24] Fan C-Y, Chen W-M, Lee OK, Huang C-K, Chiang C-C, Chen T-H. Acetabular revision arthroplasty using jumbo cups: an experience in Asia. Arch Orthop Trauma Surg 2008; 128(8): [25] Paprosky WG, Perona PG, Lawrence JM. Acetabular defect classification and surgical reconstruction in revision arthroplasty. A 6-year follow-up evaluation. J Arthroplasty 1994; 9(1): [26] Ranawat CS, Dorr LD, Inglis AE. Total hip arthroplasty in protrusio acetabuli of rheumatoid arthritis. J Bone Joint Surg Am 1980; 62(7): [27] Dou Y, Zhou Y, Tang Q, Yang D, Liu J. Leg-length discrepancy after revision hip arthroplasty: are modular stems superior? J Arthroplasty 2013; 28(4): [28] Widmer K-H. A simplified method to determine acetabular cup anteversion from plain radiographs. J Arthroplasty 2004; 19(3): [29] Delee JG, Charnley J. Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res 1976; (121): [30] Grappiolo G, Loppini M, Longo UG, Traverso F, Mazziotta G, Denaro V. Trabecular metal augments for the management of Paprosky type III defects without pelvic discontinuity. J Arthroplasty 2015; 30(6): 1-6. [31] Harris WH. Traumatic arthritis of the hip after dislocation and acetabular fractures: treatment by mold arthroplasty. An end-result study using a new method of result evaluation. J Bone Joint Surg Am 1969; 51(4): [32] Obenaus C, Winkler H, Girtler R, Huber M, Schwägerl W. Extra-large press-fit cups without screws for acetabular revision. J Arthroplasty 2003;18(3): [33] Del Gaizo DJ, Kancherla V, Sporer SM, Paprosky WG. Tantalum augments for Paprosky IIIA defects remain stable at midterm followup. Clin Orthop Relat Res 2012; 470(2): [34] Abolghasemian M, Tangsataporn S, Sternheim A, Backstein D, Safir O, Gross A. Combined trabecular metal acetabular shell and augment for acetabular revision with substantial bone loss. Bone Joint J 2013; 95(2): [35] Nehme A, Lewallen DG, Hanssen AD. Modular porous metal augments for treatment of severe acetabular bone loss during revision hip arthroplasty. Clin Orthop Relat Res 2004; 429: [36] Nwankwo CD, Ries MD. Do jumbo cups cause hip center elevation in revision THA? A 23

26 radiographic evaluation. Clin Orthop Relat Res 2014; 472(9): [37] Nwankwo C, Dong NN, Heffernan CD, Ries MD. Do Jumbo Cups Cause Hip Center Elevation in Revision THA? A Computer Simulation. Clin Orthop Relat Res 2014; 472(2): [38] Delp SL, Wixson RL, Komattu AV, Kocmond JH. How superior placement of the joint center in hip arthroplasty affects the abductor muscles. Clin Orthop Relat Res 1996; 328: [39] Bicanic G, Delimar D, Delimar M, Pecina M. Influence of the acetabular cup position on hip load during arthroplasty in hip dysplasia. Int Orthop 2009; 33(2): [40] Kelley SS, Lachiewicz PF, Hickman JM, Paterno SM. Relationship of femoral head and acetabular size to the prevalence of dislocation. Clin Orthop Relat Res 1998; 355: [41] Peter R, Lübbeke A, Stern R, Hoffmeyer P. Cup size and risk of dislocation after primary total hip arthroplasty. J Arthroplasty 2011; 26(8): [42] Lachiewicz PF, Soileau ES. Fixation, survival, and dislocation of jumbo acetabular components in revision hip arthroplasty. J Bone Joint Surg Am 2013; 95(6): [43] Odri GA, Padiolleau GB, Gouin FT. Oversized cups as a major risk factor of postoperative pain after total hip arthroplasty. J Arthroplasty 2014; 29(4): [44] Blumenfeld TJ, Meehan JP. The Use of Augment Devices in Revision Acetabular Surgery. JBJS Rev 2014; 2(3). [45] Khamis S, Carmeli E. Relationship and significance of gait deviations associated with limb length discrepancy: a systematic review. Gait Posture 2017; 57: [46] Lewinnek GE, Lewis J, Tarr R, Compere C, Zimmerman J. Dislocations after total hip-replacement arthroplasties. J Bone Joint Surg Am 1978; 60(2): [47] Patel J, Masonis J, Bourne R, Rorabeck C. The fate of cementless jumbo cups in revision hip arthroplasty. J Arthroplasty 2003; 18(2): [48] Ranawat AS, Meftah M, Thomas AO, Thippanna RK, Ranawat CS. Use of oversized highly porous cups in acetabular revision. Orthopedics 2016; 39(2): e [49] Nilsdotter A, Bremander A. Measures of hip function and symptoms: Harris Hip Score (HHS), Hip Disability and Osteoarthritis Outcome Score (HOOS), Oxford Hip Score (OHS), Lequesne Index of Severity for Osteoarthritis of the Hip (LISOH), and American Academy of Orthopedic Surgeons (AAOS) Hip and Knee Questionnaire. Arthritis Care Res 2011; 63(Suppl 11): S [50] Abolghasemian M, Tangsataporn S, Sternheim A, Backstein D, Safir O, Gross A. Porous metal augments: big hopes for big holes. Bone Joint J 2013; 95-B(11 Suppl A): [51] Babis GC, Stavropoulos NA, Sasalos G, Ochsenkuehn-Petropoulou M, Megas P. Metallosis and elevated serum levels of tantalum following failed revision hip arthroplasty--a case report. Acta Orthop 2014; 85(6):

27 Acknowledgments The authors thank Dr Ji Zhang for his help with data collection.

28 LEGEND TO FIGURES Fig. 1. Anteroposterior radiographs of three representative cases. (A-B) Two stacked slope augments (white arrow) were used inferiorly to obtain extended fixation remote to the acetabulum and to maintain the COR at an anatomic level in case 1 (note the extensive bone loss in ischium); (C-D) A trimmed shim augment (white arrow) was inserted into the ischium as a mechanically supportive point in case 2. (E-F) Slope augments (black arrow) and lotus augments (white arrow) were used to reconstruct supportive points superiorly and inferiorly, respectively. The CORs were well maintained in case 3. Fig. 2. (A) Picture of the three-dimensional printed porous titanium lotus augments composed of a disc (solid black arrow) and a stem (dotted black arrow). (B) Artistic rendering of reconstructing the inferior supportive point with a lotus augment, whose stem is inserted into the ischium and the disc supports the inferior part of the cup. Fig. 3. Radiographic measurements of biomechanical parameters. Cr = reconstructed COR; Ca = anatomic COR; AHP = absolute horizontal position of COR; AVP = absolute vertical position of COR; RHP = relative horizontal position of COR; RVP = relative vertical position of COR; FO = femoral offset; CLL = contralateral limb length; RLL = reconstructed limb length. Leg length discrepancy = RLL CLL. Fig. 4. Box and whisker plots demonstrate vertical and horizontal distances between the reconstructed COR and the anatomic COR in the Paprosky type III bone defect subgroup. (A) The vertical distance was less on average (p<0.001), with less variation (p=0.003) occurring in the metal augment group. (B) Similarly, the horizontal distance was also less on average (p=0.013), with less variation (p=0.014) in the metal augment group. Fig. 5. Box and whisker plots demonstrate vertical and horizontal distances between the

29 reconstructed COR and the anatomic COR in the subgroup of Paprosky type II bone defects. (A) Although the variations were similar between groups (p=0.079), the average vertical distance was less in the metal augment group (p=0.015). (B) The average and variation of horizontal distance did not differ between the two groups (p=0.183 and p=0.700, respectively). Fig. 6. Kaplan-Meier survival curve with radiological or clinical failure of the acetabular components as the endpoint.

30 Table 1 Comparison of Patient Demographic and Operative Variables Variables Augment Jumbo Cup p-value (n = 74) (n = 77) Gender Male 34 (45.9%) 43 (55.8%) Female 40 (54.1%) 34 (44.2%) Age (years) 58.5 ± ± * Height (cm) ± ± * Body Mass Index (kg/m 2 ) 23.5 ± ± * Reason for revision Aseptic loosening 55 (74.3%) 61 (79.2%) Infection 5 (6.8%) 6 (7.8%) Fracture 2 (2.7%) 1 (1.3%) Dislocation 6 (8.1%) 2 (2.6%) Others 6 (8.1%) 7 (9.1%) Interval between first surgery 14 (8, 18.3) 12 (6, 16.5) # and revision ** (years) Previous surgeries times ** 1 (1, 2) 1 (1, 2) # Charlson comorbidity index ** 0 (0, 1) 0 (0, 0) # Preoperative Harris Hip Score 39.3 ± ± 13.2 <0.001 * Femoral revision Stem revision 61 (82.4%) 64 (83.1%) Isolated head exchange 13 (17.6%) 13 (16.9%) Femoral head size (mm) 33.8 ± ± 2.5 <0.001 * Femoral head length ** (mm) 0 (0, 3.6) 0 (-3.5, 1.8) # Values are summarized as the incidence, with the percentage in parentheses. Chi-square test. Values are summarized as the mean ± standard deviation. * Independent samples t-test. ** Values are summarized as the median, with the 25th and 75th percentiles in the parentheses. # Mann-Whitney U test.

31 Table 2 Paprosky Acetabular Bone Defect Classification in Two Groups Classification IIA IIB IIC IIIA IIIB p-value * Augment <0.001 (n = 74) (1.4%) (10.8%) (14.8%) (25.7%) (47.3%) Jumbo cup (n = 77) (23.4%) (15.6%) (32.5%) (18.2%) (10.3%) Values are summarized as the incidence, with the percentage in parentheses. * Kruskal-Wallis test.

32 Table 3 Comparison of Biomechanical Variables Before and After revision Variables Before Revision After Revision Augment Jumbo Cup p-value Augment Jumbo Cup p-value (n = 74) (n = 77) (n = 74) (n = 77) Absolute vertical position of COR (mm) 35.9 ± ± ± ± 8.4 <0.001 Absolute horizontal position of COR (mm) 33.5 ± ± ± ± Relative vertical position of COR (mm) 19.4 ± ± ± ± 8.7 <0.001 Relative horizontal position of COR (mm) 1.2 ± ± ± ± Cup size (mm) N/M N/M N/M 58.2 ± ± 3.9 <0.001 Head-cup difference (mm) N/M N/M N/M 24.4 ± ± 3.9 <0.001 Femoral offset (mm) 32.9 ± ± ± ± Leg length discrepancy (mm) ± ± ± ± Cup inclination angle (deg) 53.1 ± ± ± ± Cup anteversion angle (deg) 15.6 ± ± ± ± Values are summarized as the mean ± standard deviation. Independent samples t-test. COR, center of rotation; N/M= not measured.

33 Table 4 Comparison of Biomechanical Variables After Revision in Paprosky Type II and III Bone Defect Subgroups Variables Paprosky Type II Bone Defect Paprosky Type III Bone Defect Augment Jumbo Cup p-value Augment Jumbo Cup p-value (n = 20) (n = 55) (n = 54) (n = 22) Absolute vertical position of COR (mm) 18.1 ± ± (13.9, 24.0) 29.8 (26.7, 34.2) <0.001 * Absolute horizontal position of COR (mm) 31.9 ± ± ± ± Relative vertical position of COR (mm) 3.3 ± ± (-0.7, 3.6) 14.1 (8.1, 19.8) <0.001 * Relative horizontal position of COR (mm) 0.1 ± ± (-8.4, 4.3) 7.9 (-8.0, 13.3) * Cup size (mm) 57.4 ± ± 3.7 < ± ± Head-cup difference (mm) 23.4 ± ± 3.9 < ± ± 4.0 <0.001 Femoral offset (mm) 35.1 ± ± ± ± Leg length discrepancy (mm) -3.9 ± ± ± ± Cup inclination angle (deg) 40.7 ± ± ± ± Cup anteversion angle (deg) 12.2 ± ± (9.9, 14.6) 10.3 (9.0, 15.2) * Values are summarized as the mean ± standard deviation. Independent samples t-test. Values are summarized as the median, with the 25th and 75th percentiles in the parentheses. * Mann-Whitney U test. COR, center of rotation.

34 Table 5 Comparison of complication and reoperation Variables Augment Jumbo Cup p-value (n = 74) (n = 77) Complication 1 (1.4%) 7 (9.1%) Aseptic loosening 0 1 (1.3%) Periprosthetic joint infection 1 (1.4%) 2 (2.6%) Dislocation 0 4 (5.2%) Reoperation 1 (1.4%) 2 (2.6%) Aseptic loosening 0 1 (1.3%) Periprosthetic joint infection 1 (1.4%) 1 (1.3%) Values are summarized as the incidence, with the percentage in parentheses. Chi-square test.

35