number.

Implants were manufactured with a highly cancellous implant surface (EPORE ®) characterized by trabeculae with a diameter of 330–390 µm to imitate trabecular bone and promote tissue ingrowth (Figure 2). –

**Figure 2.** Photograph of a finished monobloc solid body implant (Patient #2) with highly cancellous implant surface on stem, bone-facing extracortical plates, and implant body.

> The implant was designed with a hollow prosthetic body for patient #1, while a solid body was used for following implants.

> Stems or extracortical plates with supplementary interlocking screw options were used to anchor the implants to adjacent bone at implant-bone interfaces. At the proximal interface, solid stem designs were used; distally hollow stems were planned whenever feasible depending on remaining bone stock.

#### *2.5. Surgical Technique*

The stem length of tibial monobloc implants with a proximal and distal stem is a limiting factor for successful implantation. If the stem dimensions are chosen too long, they will pose an obstacle to repositioning the tibia. For this reason, the implant design of the monobloc implant of patient #4 included a distal stem with a length of only 10 mm. After resection of the distal diaphyseal tibia, axial, angular, and rotational maneuverability of the lower leg are increased even when an intact fibula remains. Therefore, in the case of patient #4, the proximal stem was implanted first while the remaining distal tibia and foot were lowered and rotated to the side as much as possible to prevent interference with the proximal implantation. When resetting the distal tibia with the distal implant interface

and stem, the existing soft tissue expansibility was used to gain the leeway necessary for implanting the 10 mm distal stem. If implantation of the distal stem had proven impossible for a lack of leeway, an additional fibular osteotomy would have been performed to gain more clearance. As the maximum amount of contrivable clearance is limited, implantation of longer implant stems would need to be planned with a modular implant design.

#### *2.6. Bone Ingrowth*

Bone ingrowth at the implant-bone interfaces was assessed clinically and radiographically. An absence of pain or instability (after full weight-bearing was achieved) served as a clinical indicator for bone ingrowth. Radiographic criteria on postoperative plain radiographs included correct implant positioning without dislocation or signs for aseptic loosening. In the event of clinical or radiographic symptoms, additional CT imaging of the reconstruction was performed.

#### *2.7. Complication Assessment*

Complications were categorized according to the Henderson classification [14].

#### *2.8. Functional Assessment*

The Musculoskeletal Tumor Society (MSTS) score and Toronto Extremity Salvage Score (TESS) were used to assess functional outcomes [15,16]. The respective questionnaires were handed out in paper form and completed by patients as part of their outpatient follow-up examinations.

#### **3. Results**

In the four patients—reconstructed using custom-made 3D-printed intercalary megaendoprostheses with a highly cancellous implant surface—distal tibia resection and belowknee amputation were avoided in all cases (Figure 3). — —

**Figure 3.** Patient #3: Reconstruction after implantation of a monobloc intercalary distal tibia implant: (**a**) intraoperative image; (**b**) postoperative radiograph of the tibia anterior-posterior (a.p.) view. L means left.

#### *3.1. Bone Ingrowth*

Primary ingrowth of the implant occurred in all patients at both implant-bone interfaces except for the proximal osteotomy line of patient #1. A partial non-union observed in this patient is more comprehensively analyzed in Section 3.3. At the current follow-up, we did not observe differences in osseointegration among treated patients (regardless of age at operation).

#### *3.2. Soft Tissue Ingrowth*

A complete ingrowth of muscular tissue into the highly cancellous implant surface was confirmed in one patient who underwent a revision for partial non-union (Figure 4).

(**a**) (**b**)

**Figure 4.** Patient #1: Intraoperative images during revision operation nine months after primary reconstruction: (**a**) view of proximal implant-tibia interface with visibility of partial non-union and soft tissue ingrowth into the highly cancellous implant surface; (**b**) complete ingrowth of muscles and soft tissues into the highly cancellous implant body surface.

#### *3.3. Complications*

— The first patient reconstructed using a highly cancellous 3D-printed monobloc intercalary tibia implant developed an incomplete non-union at the proximal bone-implant interface (Henderson Type 3—structural failure). Two extracortical plates with supplementary interlocking screws bridging the bone-implant interface were used to anchor the implant to the proximal tibial diaphysis without a central stem. This anchorage design was chosen to allow filling the hollow implant body with autologous iliac crest graft (Figure 5). Operative revision and additional plating of the bone implant interface while retaining the original implant were performed 9 months after primary reconstruction (Figure 6). In addition, the ipsilateral fibula was osteotomized and fixed to the tibial column using screw osteosyntheses after roughening the facing bone cortices to encourage bone union. A hypertrophic pseudarthrosis recurred at the tibial bone-implant interface while the fibular transfer consolidated and continues to stabilize the reconstruction by taking part of the load. The patient currently has full weight bearing using a light brace and declines further operative revision as her activities of daily life are not impaired and she has no athletic ambitions. Implant design has been adapted to include a central stem and solid implant body at the proximal bone-implant interface. After this alteration, non-union and hypertrophic pseudarthrosis were avoided in later patients.

**Figure 5.** Patient #1: Computed tomography scan 43 months after primary reconstruction: (**a**,**b**) coronar view of the implant with the depiction of the bone graft-loaded hollow implant cavity and persistent non-union of the proximal implant-bone interface.

**Figure 6.** Patient #1: Postoperative radiographs after osteosynthetic plating of the proximal implant-bone interface (**a**) a.p. view; (**b**) lateral view. R means right.

Soft tissue failure (Henderson type 1), aseptic loosening (Henderson type 2), periprosthetic infections (Henderson type 4), or local recurrence (Henderson type 5) were not observed in this collective.

#### *3.4. Functional Outcome*

All patients have achieved full weight-bearing and returned to their activities of daily life. The mean MSTS and TESS scores were 23.5 and 88, respectively. Patient #4 completed the functional questionnaire three months after the operation when presenting at the outpatient clinic for her first postoperative follow-up. She has not completed a functional rehabilitation program due to ongoing adjuvant chemotherapy yet.

#### **4. Discussion**

In this study, we present the short- to intermediate-term outcomes of four patients reconstructed using 3D-printed patient-individualized intercalary tibia monobloc prostheses with highly cancellous implant surfaces. From our point of view, complete ingrowth of soft tissues into the highly cancellous implant surface and continued joint salvage of the ankle joint despite little remaining bone stock are the most significant findings of this study.

The rationale for using titanium aluminum vanadium alloy (TiAl6V4) implants for the presented implant design were based on two main considerations: biocompability and choice of available production technique. Titanium alloys are a certified and reliable material with good biocompability for non-cemented reconstructions. They also have sufficient stability and processed using EBM allows for highly porous implant surfaces.

Periprosthetic infection is a serious problem affecting primary and revision total joint arthroplasties [17,18], but infection rates of megaendoprosthetic reconstructions are even higher despite implant features such as silver coating [19]. Possible reasons are larger reconstruction lengths with larger implant surfaces, longer operation times, and frequently immunocompromised patients. McConoughey et al., report that bacteria are often introduced into the wound in their planktonic form, growing in joint fluids before colonizing an implant. Later, they form biofilms to avoid exposure to high doses of antibiotics, develop resistance, and persist in a dormant state [20].

Complete ingrowth of soft tissues into the implant surface, as observed in this study, addresses many known causes and promotive factors in the development of periprosthetic infection: reduction of (a) dead space, (b) scar tissue formation surrounding the implants at a distance, and (c) excessive joint fluid formation around the implant.

In a study by Cordero et al., rough titanium alloy surfaces have shown a higher tendency of bacterial colonization when compared with smooth surfaces [20,21]. However, soft tissue ingrowth and accessibility of implant surfaces for immune cells may balance this observed disadvantage. While we concede that a lack of periprosthetic infection in the presented case series is not sufficient to make any firm conclusions about the impact of soft tissue ingrowth on implant surfaces, highly cancellous implant surfaces seem a feasible implant modification warranting further research.

Bone and soft tissue ingrowth also have implications for functional outcomes. MSTS and TESS scores of 23.5 and 88 in this study were satisfactory and most likely caused by the preservation of the ankle joint as well as soft tissue ingrowth. They compete with or exceed functional outcomes observed after biological and endoprosthetic intercalary or osteoarticular distal tibia reconstructions. Tanaka et al. reported MSTS scores of nineteen patients ranging between 93–100% after reconstruction with vascularized fibula grafts in intercalary femur and tibia defects. They also observed a union rate of 79% after a mean time of 7.8 months, which necessitated long periods of partial to no weight-bearing [22]. Khira et al., published a mean MSTS score of 84 (80–92) in their collective of patients reconstructed using vascularized fibula grafts with an Ilizarov external fixator for large tibial bone defects [23]. Intercalary or osteoarticular distal tibia allografts (optionally augmented with vascularized fibula grafts or composite prostheses) are another biological reconstruction option reported by Donati et al. However, they were rarely indicated for the ankle joint compared with other locations (*n* = 3). Complications included non-union (49%) and fracture (27%) observed in all reconstructed sites (*n* = 112) [24].

Abudu et al. reported their results for endoprosthetic replacement of the distal tibia and ankle joint (*n* = 5). While function was excellent to begin with, it deteriorated over time. Yet, patients maintained a mean Enneking score of 50% in this study [5]. In 2017, Yang et al. documented a median MSTS score of 66% in eight patients treated by custommade distal tibia megaprosthesis [2]. Lee et al. assessed a mean MSTS score of 24.2 in six patients treated with the custom-made, hinged distal tibia and ankle prostheses. Among

complications, talar collapse and wound infection were noted [6]. Shekkeris et al. reported that two of six patients treated by endoprosthetic distal tibia replacement went on to have below-knee amputation for persistent infection after a mean of 16 months in their study. The mean MSTS and TESS score of patients retaining the implants was 70% and 71% [3]. The most common complication after endoprosthetic intercalary reconstruction observed by Streitbürger et al. was aseptic loosening. Alteration of stem design to fit biomechanical demands of epi- or metaphyseal stem anchorage showed a tendency to improve implant longevity, though. [10].

The complication rates presented in the above-mentioned studies prove that reconstruction of bone defects of the distal tibia after tumor resections remains challenging regardless of the reconstruction technique chosen. Furthermore, the authors agree that below-knee amputation remains a valid treatment choice. The implant design presented in this case series achieved joint and limb salvage at a low complication rate and satisfactory functional outcomes with an early return to full weight-bearing. For this reason, increased consideration of biomechanical demands on implants and further technological advancements of 3D-printing seem a promising research avenue to increase the role of megaendoprosthetic reconstructions in this challenging location.

#### **5. Conclusions**

Reconstruction of the diaphyseal and distal tibia using custom-made 3D-printed intercalary implants proves to be a feasible treatment strategy in this case series. Considering that functional outcome after below-knee amputation leads to acceptable results and lower complication rates compared with other limb salvaging biological and standard megaendoprosthetic approaches, a low complication rate and good functional outcome in this case series emphasize that limb salvage using 3D-printed custom-made implants should be considered when counseling patients despite a lack of long-term experiences. However, observation of these patients with regard to long-term functional results and complication rates is necessary. Highly cancellous implant body designs should be considered in other megaendoprosthetic implant sites as well and future studies investigating this design's advantages and complications seem warranted.

**Author Contributions:** Conceptualization, W.K.G., J.H. and A.S.; methodology, not applicable; software, not applicable; validation, W.K.G., J.H. and A.S.; formal analysis, W.K.G.; investigation, W.K.G.; resources, W.K.G.; data curation, W.K.G.; writing—original draft preparation, W.K.G.; writing—review and editing, W.K.G., J.H., M.N., L.E.P. and A.S.; visualization, W.K.G.; supervision, not applicable; project administration, W.K.G.; funding acquisition, not applicable. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Committee of the University of Duisburg-Essen (protocol code 21-9859-BO, approved 16 February 2021).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available on reasonable request from the corresponding author.

**Conflicts of Interest:** The authors W.G., M.N. and L.P. declare no conflict of interest. The authors J.H. and A.S. received research grants from "implantcast" company and financial support for attending symposia. The company "implantcast" had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


## *Article* **Custom Made Monoflange Acetabular Components for the Treatment of Paprosky Type III Defects**

**Sebastian Philipp von Hertzberg-Boelch 1, \* , Mike Wagenbrenner 1 , Jörg Arnholdt 1 , Stephan Frenzel 2 , Boris Michael Holzapfel <sup>1</sup> and Maximilian Rudert 1**


**Abstract:** Purpose: Patient-specific, flanged acetabular components are used for the treatment of Paprosky type III defects during revision total hip arthroplasty (THA). This monocentric retrospective cohort study analyzes the outcome of patients treated with custom made monoflanged acetabular components (CMACs) with intra- and extramedullary iliac fixation. Methods: 14 patients were included who underwent revision THA with CMACs for the treatment of Paprosky type III defects. Mechanism of THA failure was infection in 4 and aseptic loosening in 10 patients. Seven patients underwent no previous revision, the other seven patients underwent three or more previous revisions. Results: At a mean follow-up of 35.4 months (14–94), the revision rate of the implant was 28.3%. Additionally, one perioperative dislocation and one superficial wound infection occurred. At one year postoperatively, we found a significant improvement of the Western Ontario and McMaster Universities Arthritis Index (WOMAC) score (*p* = 0.015). Postoperative radiographic analysis revealed good hip joint reconstruction with a mean leg length discrepancy of 3 mm (−8–20), a mean lateralization of the horizontal hip center of rotation of 8 mm (−8–35), and a mean proximalization of the vertical hip center of rotation of 6 mm (13–26). Radiolucency lines were present in 30%. Conclusion: CMACs can be considered an option for the treatment of acetabular bone loss in revision THA. Iliac intraand extramedullary fixation allows soft tissue-adjusted hip joint reconstruction and improves hip function. However, failure rates are high, with periprosthetic infection being the main threat to successful outcome.

**Keywords:** patient specific implant; custom made implant; revision hip; Paprosky; pelvic discontinuity

#### **1. Introduction**

The revision burden after total hip arthroplasty (THA) will increase [1]. Acetabular bone loss is a major surgical challenge in revision THA (rTHA), particularly in re-revisions or after implant migration. Successful acetabular reconstruction with long-term component fixation requires sufficient primary stability for secondary osteointegration. A broad range of surgical strategies are available, of which the most popular are antiprotrusion cages [2], hemispherical or asymmetrical cups with intra- or extramedullary fixation [3–5] and modular, highly porous acetabular revision systems with and without metal wedges, buttress augments, and cage options [6,7]. However, it has not yet been defined which strategy should be considered as the benchmark [8].

Although custom made implants consume great organizational and financial resources, they are a further treatment option for large osseous defects that otherwise cannot be managed with standard implants. Based on computed tomography (CT), custom made acetabular components offer the surgeon the option to add metal sockets to the implant

**Citation:** von Hertzberg-Boelch, S.P.; Wagenbrenner, M.; Arnholdt, J.; Frenzel, S.; Holzapfel, B.M.; Rudert, M. Custom Made Monoflange Acetabular Components for the Treatment of Paprosky Type III Defects. *J. Pers. Med.* **2021**, *11*, 283. https://doi.org/10.3390/jpm11040283

Academic Editor: Klaus Radermacher

Received: 1 March 2021 Accepted: 6 April 2021 Published: 8 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

volume according to the defect of the hemipelvis, to adjust flanges for fixation devices to the remaining bone stock, and to plan the reconstruction of the hip center of rotation (COR) [9].

Most custom made acetabular components are designed as triflanges. These custom made triflange acetabular components (CTACs) are intended to "span the gap" by bridging the periacetabular defect and provide fixation options at the Os ilium, the Os pubis, and the Os ischium. However, these components were initially designed for the posterior approach, and the approach has to be relatively extensile to position all three flanges correctly. Consequently, results for these acetabular implants are highly variable [10].

At the study institution, high-grade acetabular bone defects are treated with different types of "off the shelf" acetabular reconstruction systems via the anterior but mainly the lateral or anterolateral approach. One of these systems relies on the combination of extraand intramedullary iliac fixation using an iliac flange and an optional intramedullary press-fit stem and has proven good results in various studies [4,5]. However, there are defect situations in which "off the shelf implants" do not seem to be appropriate, for instance, in cases with significant loss of supportive bone from the anterior or posterior acetabular rim maybe with additional resorption of the dome. For these patients, a custom made monoflanged acetabular component (CMAC) seems warranted. The iliac flange is fixed to the gluteal surface of the ilium and can be positioned via the standard anterior or lateral approaches. The implant can be armed with an intramedullary press-fit stem for additional fixation.

In the following, we report on the patients who have been treated with this implant for reconstruction of the acetabulum after complicated rTHA.

#### **2. Methods**

#### *2.1. Patient Selection*

Approval for this retrospective study was given by the institution's review board (Reference number 2016072801). We retrospectively identified 18 cases that underwent acetabular reconstruction with CMACs between January 2010 and December 2019 at our department. Three cases were excluded since the indication for CMAC was malignancy, and one patient died during CMAC implantation.

#### *2.2. Implant*

The CMAC is designed using data obtained via high-resolution CT imaging of the pelvis with an implant-specific algorithm (WinCad, Fa. AQ Solutions). Scans can be performed with or without a prosthesis or spacer in place. Figure 1 illustrates crucial templating steps that can be modified by the surgeon. After design approval by the surgeon, the implant is produced by laser melting of a titanium alloy (TiAI6V4) in a monoblock fashion with a 3D comb surface structure and with optional HA or CAP layering. The variability in form is reflected by Figures 1, 3 and 4. Manufacturing and delivering takes about 6 to 8 weeks.

#### *2.3. Parameters Assessed*

All presented data were extracted from the electronic patient charts. Preoperative acetabular defect situation was classified according to the modified Paprosky classification system based on preoperative radiographs and CT scans as described previously [11]. Postoperative radiographs were evaluated for reconstruction of the hip joint's COR according to Rannawat [12]. Leg length discrepancy (LLD) was assessed by comparing the position of the trochanter minores to the connection line between Kohler's teardrops.

**Figure 1.** Selected templating steps for a custom made monoflanged acetabular component (CMAC) with optional stem for intra- and extramedullary iliac fixation for a Paprosky IIIA defect: (**A**) Assessment and 3D visualization of the defect situation with and without subtraction of the implant. (**B**) CT-based estimation of leg length discrepancy (LLD) respecting pelvic tilt and joint contractures. (**C**) Virtual reconstruction of the hip center of rotation (COR) by positioning a standard acetabular component of a specific size at the anatomical COR. Bone that has to be reamed to position the original implant is colored in red. (**D**) Design features of the CMAC: The large segmental iliac defect is filled by the implant's metallic monoblock assembled socket. Screws are positioned in areas of the pelvis that are characterized by intact host bone with a recommendation for their length in millimeters. For further primary stability, the surgeon can implant an additional intramedullary press-fit stem (entrance point colored in red). Planning and defect classification have previously been described in detail by our group (11).

Perioperative complications were defined as complication within 3 months after CMAC implantation and were tabulated as documented in the electronic patient chart. Implant revision after CMAC implantation was considered a failure. Failures were excluded from functional follow-up. Functional outcome was assessed using the Western Ontario and McMaster Universities Arthritis Index (WOMAC) Score that was recorded prospectively before and one year after CMAC implantation. Latest ap pelvic radiographs were examined for signs of implant loosening. Therefore, radiolucency lines thicker than 2 mm with sclerotic demarcation were considered significant [13]. Since the Charnley and DeLee zones are not applicable, 4 zones around the implant were defined: the cup, the metal socket, the iliac stem, and the iliac flange.

#### *2.4. Patients*

The cohort consisted of 14 patients, 5 men and 9 women. The operations were performed by hip and knee arthroplasty surgeons with additional specialization in revision cases. The operating surgeon indicated treatment with CMAC. Major decision criterion was bone loss at the ilium that did not enable adequate hip COR reconstruction with "off the shelf" cup and cage constructs or asymmetrical cups with intra- or extramedullary fixation. The mean age was 69.5 years (55–83), and the mean body mass index (BMI) was 28.0 kg/m<sup>2</sup> (24.5–30.9), respectively. A total of 11 patients were classified as ASA III, and 3 patients as ASA II. Seven patients had no previous rTHA, the other 7 patients underwent 3 or more previous revisions. Indications for CMACs were spacer implantation after infection in 4 and aseptic loosening in 10 patients.

#### *2.5. Statistics*

Parameters are shown as mean and range. A Kaplan–Meier analysis for revision-free survival was performed. The Wilcoxon test was used to test paired samples for significance. A *p*-value <0.05 was assumed significant. Statistics were conducted with SPSS.

#### **3. Results**

Characterization of treated acetabular defects and treatment strategy is shown in Table 1. No additional osteosynthesis at the pelvis was performed during CMAC implantation.


**Table 1.** Classification of acetabular defects with treatment strategy and failures (number = N; pelvic discontinuity = PD).

#### *3.1. Intraoperative Parameters*

All patients were operated in supine position. A transgluteal Bauer approach was used in all but two patients, for which an anterolateral approach was more suitable. A semiconstrained liner was cemented into the acetabular construct in all but two patients who received a standard liner. Five patients underwent complete THA removal and spacer implantation before proceeding to CMAC implantation. The mean operation time for the seven patients with additional femoral stem exchange was 181 min (107–249). The mean operation time for the seven patients with only acetabular component exchange was 175 min (93–243). Postoperative weight bearing was restricted for 6 weeks in 11 patients and for 12 weeks in the remaining 3.

#### *3.2. Perioperative Complications*

One patient suffered from postoperative dislocation, which was managed with closed reduction. Another patient had a superficial wound infection that was managed with debridement. We did not observe perioperative fracture, nerve injury, or deep vein thrombosis.

#### *3.3. Failures*

The mean follow-up was 35.4 months (14–94). We observed two acute septic failures (14.3%) at 10 and 35 months after CMAC implantation, which were treated with debridement, antibiotic therapy, irrigation, and implant retention. However, moving parts were exchanged in these two cases. Further, we observed two aseptic CMAC loosenings (14.3%). One patient was converted to a jumbo head after 14 months, and the other revised and the acetabular component replaced with a modular revision system 20 months after CMAC implantation. Figure 2 depicts the cumulative revision-free survival.

**Figure 2.** Kaplan–Meyer estimate of revision free-survival.

#### *3.4. Function*

Table 2 shows significant improvement of the WOMAC score and its subgroups pain and physical function in patients without failure one year after CMAC implantation. One patient did not complete the WOMAC questionnaires completely.

**Table 2.** Patient-reported function assessment with the Western Ontario and McMaster Universities Arthritis Index (WOMAC) score (number = N); \* only complete pairs were included.


#### *3.5. Radiographic Evaluation*

Results of radiographic evaluation are shown in Table 3. Two patients were planned with intentional extra-anatomic reconstruction of COR as depicted in Figures 3 and 4.

> − −

**Table 3.** Radiographic evaluation.


**Figure 3.** Radiolucency lines without need for revision: (**A**) Preoperative radiographic situation showed the acetabular "up-and-out" defect after implantation of a large head because of acetabular bone loss. (**B**) Anterior to posterior (left) and posterior to anterior (right) views show the intended proximalization of the COR in the virtual 3D reconstruction. The cup is not placed at the level of the Kohler's tear drop. (**C**) Radiograph after revision and CMAC implantation showed restoration of leg length with a high COR. (**D**) Significant radiolucency lines developed around the whole implant at 3 years of follow-up. Although implant migration cannot be excluded, the patient was not revised because he reported daily walks of up to 10 km supported by a cane. Thus, this case was not considered a failure.

For the 10 patients without failure, significant radiolucency lines were found around the socket for one patient, around the acetabular construct for another patient, and around the whole CMAC for the last (Figure 3):

**Figure 4.** Osteointegration of the socket at follow-up. (**A**) Preoperative situation demonstrated an "up and out" defect that was filled by the loosened cup and augment construct. (**B**) The radiographic control after 2 years displayed PD with complete disruption of the ilio-ischial line and medial protrusion of the cup. (**C**) Radiographic situation 2 years after revision showed no sign of loosening. (**D**) In the CT, spot welds, as sign of osteointegration at the HA-coated socket, were seen.

#### **4. Discussion**

Acetabular bone loss remains a major surgical challenge in complicated rTHA. With the presented CMAC we found acceptable results with significant improvement of function one year after implantation and an implant revision rate of 28.6% at a mean follow-up of 35.4 months.

The reported outcome is certainly influenced by patient-related presuppositions for acetabular reconstruction, which are anteceding or even subliminal infection and massive bone loss. In the current study, all patients had at least a Paprosky type III acetabular defect and 42.86% of patients even displayed PD. The optimal surgical strategy for those patients has not yet been defined. A stable pelvic ring is discussed as the "conditio sine qua non" for prevention of mechanical failure of acetabular constructs [7]. Antiprotrusion cages and CTACS as well as cup cage constructs aim to fulfill this strategy [14]. In contrast, the presented implant design abandons this strategy and relies on a combination of intraand extramedullary iliac fixation for primary stability. However, positioning of the stem can be challenging. In two cases with a IIIa defect but with sufficient medial abutment by the remaining bone, the stems were dispensed. Implant loosening was not observed in these cases. However, whenever possible the iliac stem should by applied for optimal fixation. A rigid fixation of the CMAC to the Os ilium allows osteointegration as depicted in postoperative CT scans (Figure 4).

Irrespectively of the fixation strategy, component fixation seems to be rather successful while other complications are frequent. This statement is underlined by the literature and the data presented data here with high complication rates but acceptable acetabular component survival: In the review of CTACs by Chiarlone et al., the acetabular component survival rate ranged from 86.5% to 100%, but the reoperation rate was 24.5% [15]. In the review by De Martino et al., aseptic loosening of CTACs occurred in only 1.7%. However, the complication rate was 29% [10]. CTACs are designed to span the periacetabular gap by fixation to the iliac, the ischial, and the pubic bone. In contrast, Burastero et al. described a modular press-fit implant design with an antiprotrusion collar for patientspecific acetabular reconstruction and observed osteointegration of all implants at followup [8]. The acetabular component survival rate in the current study was 85.72%. However, overall complications occurred in 42.86%. This extremely high rate is comparable to the rate reported in the literature. De Martino et al. and Chiarlone et al. reported reoperation and complication rates ranging from 0 to 66.7% [10,15].

To the best of our knowledge, there is only one other study analyzing the outcome of CMACs. Walter et al. investigated and compared the outcomes of different designs of CTACs. With a mean follow-up of 79.8 months for the CTAC group and 43.0 months in the CMAG group, they found no significant difference regarding the implant survival rate, which was 28.6% and 21.6%, respectively [16].

In comparison to the three-point fixation for CTACs, the iliac fixation is advantageous because it requires less preparation at the ischium. Additionally, it can routinely be implanted in supine position, which facilitates leg length evaluation.

There are limitations to this study. Acetabular defects were assessed based on the preoperative templating CTs, instead of radiographs as initially described by Paprosky. This is warranted for the following reasons: First, radiographic evaluation is not feasible if the volume of the indwelling prosthesis covers bony landmarks. Second, PD does not always match the Paprosky classification [7] and finally, radiographic evaluation has demonstrated high inter- and intraobserver variability and tends to underestimate the acetabular defect situation [11,17]. However, it remains the most popular classification system for acetabular bone loss.

The mean follow-up reflects only the short-term outcome, and the number of patients is limited. Because this study focused on CMACs as revision implants, three patients were excluded due to malignancy. Although the revision burden is increasing, patients that do not meet the criteria for the treatment with an "off the shelf" acetabular revision system are still rare. This limitation is reflected by the overall small number of only 579 and 627 patients in the aforementioned reviews [10,15]. To the best of our knowledge, the current study reports the largest cohort study of patients treated for acetabular bone loss after rTHA failure with one special CMAC design.

Due to the retrospective design of this study, we cannot directly compare the results to those of CTACs. In our hands, the advantages of monoflange fixation are so convincing that we prefer it over the use of CTACs. However, we do observe a trend to highly porous cup-cage constructs with optional wedges and buttress augments. This is mainly based on the instant availability and intraoperative flexibility. However, the surgical strategy used and its success is still highly dependent on the surgeon's skills and his/her experience with a particular implant. CMACs should be considered in cases with a high-grade acetabular defect situation, in which particularly cranial or caudal acetabular bone loss endangers successful reconstruction.

#### **5. Conclusions**

CMACs can be considered an option for the treatment of acetabular bone loss in rTHA. Preoperative CT-based 3D planning yields reproducible results for leg length and hip COR. The limited available data show that iliac intra- and extramedullary fixation allows soft tissue-adjusted hip joint reconstruction and improves hip function. However, failure rates are high with periprosthetic infection being the main threat to successful outcome.

**Author Contributions:** All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by M.W., S.P.v.H.-B. and J.A. The first draft of the manuscript was written by S.P.v.H.-B. and all authors commented on previous versions of the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This publication was supported by the Open Access Publication Fund of the University of Wuerzburg.

**Institutional Review Board Statement:** Approval for this retrospective study was given by the institution's review board (Reference number 2016072801).

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

**Conflicts of Interest:** The authors declare that they have no known competing financial interest or personal relationships that could have appeared to influence the work reported in this paper.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Journal of Personalized Medicine* Editorial Office E-mail: jpm@mdpi.com www.mdpi.com/journal/jpm

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com ISBN 978-3-0365-4247-8