Intracavitary and Extracavitary Iliac Extended Fixation Using Porous Metal Augments for Paprosky Type 3 Acetabular Defects in Revision Total Hip Arthroplasty: A Minimum 2 Years Follow-up

Background: We proposed a new concept of “iliac extended fixation” in revision total hip arthroplasty (THA) as fixation extending superiorly 2 cm beyond the original acetabular rim with porous metal augments, which was further classified into intracavitary and extracavitary fixation. This study reports the minimum 2 years outcome of the iliac extended fixation technique in patients with Paprosky type 3 acetabular defects. Methods: Thirty two revision THA patients were retrospectively reviewed who underwent reconstruction with the concept of iliac extended fixation from 2014 to 2016 in our hospital. Patients were assessed using the Harris Hip Score (HHS) and the Western Ontario and McMaster Universities Osteoarthritis Index score (WOMAC). In addition, radiographs were assessed and patient reported satisfaction was collected. Results: At an average follow-up of 40 months (range 25 – 64 months), the postoperative HHS and WOMAC scores were significantly improved at the last follow-up (p < 0.001). The postoperative horizontal and vertical locations of the COR were significantly improved from the preoperative measurements (p < 0.05). Two (6.3 %) patients were dissatisfied with the outcome. The extracavitary iliac extended fixation group had significantly more horizontal screws fixation (4.6 ± 25.7° vs 41.9 ± 37.8°, p < 0.001) and higher rate of osteointegration in zone 1A (91.7% vs 50.0%, p = 0.023) than the intracavitary iliac extended fixation group. Conclusions: Intracavitary and extracavitary iliac extended fixation with porous metal augments and cementless cups are effective in reconstructing severe superior acetabular bone defects, with promising short-term clinical and radiographic outcome.


Introduction
The quality of the superior acetabular bone stock is critical to prosthetic acetabular component fixation and load transmission [1]. Unfortunately, superior acetabular bone defects are commonly encountered in revision total hip arthroplasty (THA) [2][3][4][5]. Paprosky type 3 defects indicate severe and extensive superior acetabular bone loss, often requiring reconstruction to be extended to an area at least 2 cm above the original acetabular rim [2]. Several different methods exist as a means of reconstructing superior bone defects, including jumbo cups, structural allografts, cup-cage construction, and impaction bone graft with cemented cups [4,5]. Each method has been successful in selected cases. However, these traditional methods frequently encountered relatively lower survivorship due to mechanical failures, or compromised hip biomechanics as a result of highly riding hip centers resulting from obtaining adequate bony coverage of cup, particularly in patients with severe superior acetabular bone defects [6,7].
The introduction of porous metal augments has provided a promising solution to this problem [8][9][10][11], and have been used in the reconstruction of Paprosky type 3 severe superior acetabular bone defects. Jenkins et al. classified the use of metal augments in the superior aspect of the acetabulum into the three types, namely "flying buttress", "dome" and "footing" [12]. This classification is according to the augment's location and orientation, and is instructive to the intraoperative application of porous metal augments. However, it does not cover all types of intraoperative applications of porous metal augments, especially those extending beyond the inferior pole of the sacroiliac joint, like buttress augments and stacked slope augments, which have the potential to feature distinct technical and biomechanical characteristics. There is a paucity of literature reporting the intraoperative classifications and rationale for reconstruction using porous metal augments.
In the current study, we propose the concept of "iliac extended fixation", as compared to the jumbo cup technique and juxta acetabular fixation, which are commonly used in patients with Paprosky type II defects. Iliac extended fixation is defined as acetabular reconstruction components that are fixed to the iliac bone at a level 2 cm higher than the original superior edge of the acetabulum. The rational for the 2 cm cutoff is based on two aspects: on one hand, our clinical observation that in most patients, the bone structure turns to the ala of ilium, where bone stock is dramatically decreased compared with that of the juxta-acetabular area within 2 cm of the original acetabular rim [13]; and on the other hand the biomechanical disadvantages of increased joint reaction force, risk of loosening, and decreased range of motion by migrating center of rotation superiorly over 2 cm [14][15][16][17][18][19][20].
Utilizing cementless porous metal augments, iliac extended fixation is further classified into intracavitary and extracavitary fixation types, which encompass different local mechanical conditions of compressive and shear stresses, respectively. Intracavitary iliac extended fixation (IIEF) was defined as cavitary bone loss exceeding 2 cm superior to the original acetabular rim, and a single slope augment or stacked augments with ≥ 2cm thickness that could be contained in the bone bed and stabilized by compressive stress with minimal shear stress.
Extracavitary iliac extended fixation (EIEF) was defined as the segmental superior bone loss so extensive that augments could not be contained in the bone bed. Thus, achievement of primary stability in EIEF required directly fixing an augment to the iliac wing using multiple buttress screws to resist shear stress. Both IIEF and EIEF sought to provide stable superior points of fixation with a well-positioned cementless hemispherical cup.
This new concept may help establish intraoperative mechanical principles to guide the reconstruction of severe superior acetabular bone defects. With the IIEF and EIEF techniques, load is transmitted to the superior area distant from the original acetabular rim, with the shear and compressive stresses taken into consideration in the choice and application of porous metal augments during revision THA. In the current study, we asked two major questions: ① What are the minimum 2 years postoperative clinical and radiological results of iliac extended fixation technique in patients with Paprosky type 3 superior acetabular defects? ② Are there any preoperative, intraoperative, and postoperative differences between the two groups of IIEF and EIEF techniques?

Patients
This study was approved by the hospital ethics committee. In order to enroll patients reconstructed with the concept of iliac extended fixation, 102 consecutive patients who underwent revision THA with the use of porous metal augments at our institution from January 2014 to October 2016 were retrospectively reviewed. The inclusion criteria consisted of patients undergoing revision THA, acetabular reconstruction with a cementless hemispherical cup and porous metal augments, construction extending superiorly to more than 2 cm above the original upper acetabular rim, and minimal follow-up of 24 months. Of the initial 99 patients that were reviewed, 32 patients were deemed eligible based on the inclusion criteria and were therefore enrolled in this study. The other 72 patients were excluded because the augments did not extend over 2 cm above the original acetabular rim. Trabecular metal augments made from tantalum (TM Augment; Zimmer, Warsaw, IN) or titanium (Restoration wedge augment; Stryker, Mahwah, NJ) were utilized to reconstruct the superior acetabular defects. There were twelve patients reconstructed with the EIEF technique, using a single buttress augment or combined with other augments, or a flying buttress using slope augments [12] (Table 1); twenty patients reconstructed with the IIEF technique, fourteen of whom using a single superior slope augment, and the other six with stacked superior slope augments ( Table 2) The average patient age at the time of surgery was 57 years (range 38 -78 years), the average body mass index (BMI) was 25.0 (range 18.7 -33.7), and 13 (40.6 %) patients were female. On average, patients had 1.6 (range 1 -5) previous hip operations. There were 21 patients (65.6%) revised for aseptic loosening, 6 (18.8%) for periprosthetic infection, 3 for recurrent dislocation (9.4 %), 1(3.1%) for severe acetabular side osteolysis without component loosening, and 1 (3.1%) for periprosthetic fracture. The preoperative Charlson comorbidity score was 0 in 26 (81.3%) patients, 1 in 4 (12.5%) patients, and 2 in 2 patients (6.3%). The average duration from first THA to surgery was 14 (range 0.5-31) years. The average number of previous surgeries for the index hip was 1.6 (range 1-5).

Surgical Technique
All the patients were cleared of infection at the time of index revision surgery.
Patients revised for infection were treated with staged revision and sensitive antibiotics for 6 weeks, and the final revisions were not commenced until joint aspirations as well as ESR and CRP turned normal. Same protocol was followed for the patient with acute infection after revision. A posterior lateral approach was employed in all hips. The femoral head was removed after dislocation of the hip, and the femoral stem was removed if it was deemed loose or considered necessary for exposure, reconstruction, or if evidence of trunionosis was visible. The femoral stems in 7 (21.9%) hips were preserved with only the acetabular side revised.
Extended trochanteric osteotomy was done in 6 (18.8%) hips because of difficulty in removing the femoral components.
The acetabular cups were then removed, and intraoperative tissues were sent to culture and pathology for detection of any possible residual infection. Fibrous tissue was then removed to fully expose the acetabulum and the bone defects. Patients with pelvic discontinuity were then treated with distraction technique by inserting a 3 mm pin into the ilium and ischium ramus, respectively, and then distracting the cranial and caudal fragments with a pliers holding the two pins, in order to stabilize the anterior and posterior column before preparation of augment and cup bony bed [21,22].
The site, size, and shape of the bone defects were determined before reaming and preparing for insertion of the augment and metal shell. A hemispherical cup trial or reamer was positioned to a target level by aiming at the optimal COR and contact area of the bone bed. Augment trials were then used to determine the shape, size and position of the final metal augment.
Buttress and slope augments were implanted first and fixated with screws to the host bone, followed by a multiple holed, trabecular metal shell that was impacted and press fit to ensure primary stability of the whole cementless construct. For a slope augment, a tantalum augment (TM Augment; Zimmer, Warsaw, IN) was the first choice for use; if the available bone bed required more versatile screw fixation, a titanium augment (Restoration Wedge Augment; Stryker, Mahwah, NJ) was considered (Figures 1 and 2). Cement was placed between the augments and the augment-cup interface to unify these metal components into one solid unit. Multiple screws were then inserted through the porous metal shell. Polyethylene liners with a 10° high wall were used in 14 (43.8%) patients, and neutral polyethylene liners in the other half of patients. A 28 mm diameter Co-Cr-Mo alloy head was used in 2 (6.3%) patients, and 32 mm, 36 mm and 40 mm diameter ceramic heads in 18 (56.3%), 11 (34.4%) and 1 (3.1%) patients, respectively. There were no significant differences in the preoperative demographic factors between the EIEF and IIEF groups (Table. 4). The EIEF group had significantly higher preoperative vertical location of COR, and more screws used for augment fixation than the IIEF group (Table. 4).
Postoperatively, patients with IIEF were generally allowed to ambulate with touchdown weight bearing for the first 4 weeks, then encouraged to walk with partial weight bearing for another 8 weeks, which could be accelerated according to the intraoperative evaluation of reconstruct stability by the senior author (YXZ).
Patients with EIEF were allowed to start touch-down weight bearing 2 weeks after surgery and for 6 weeks, and then partial weight bearing for another 4 weeks. By the 12th week, all patients were urged to walk with full weight bearing. Antibiotic prophylaxis with 2 grams of cefuroxime every 12 hours was initiated and then discontinued 48 hours post-operatively for aseptic cases. Antibiotics sensitive to the specific bacteria were used for second stage revision of infection cases according to previous antimicrobial susceptibility testing. Subcutaneous heparin and intermittent elastic compression were used as routine measures to prevent thromboembolic complications in the hospital, which was transitioned to low dose aspirin (100 mg per day) for 5 weeks after discharge.

Clinical Data
Patients were asked to revisit the hospital at 3 months, 6 months, 1 year, and then yearly after surgery. The Harris Hip Score (HHS) and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC, range 0-96 with higher score indicating worse outcome) were used for clinical evaluation. Clinical failure was defined as loosening or any complications requiring surgical removal of the cup.
Patients also reported satisfaction based on an five-level subjective scale (very satisfied, satisfied, neutral, dissatisfied, or very dissatisfied), which was collected at the last follow-up [23].

Radiological Evaluation
Anteroposterior, lateral and Judet view radiographs of the bilateral hips were taken for all patients 1 week prior to the planned operation. Two surgeons (YH, HT) assessed the severity of the acetabular bone deficiency according to the preoperative radiographs on the basis of the Paprosky acetabular bone defect classification [24,25]. Any discrepancies between the two surgeons were reexamined by the senior consultant surgeon (YXZ). There were 13 (40.6 %) patients classified as Paprosky type 3A, 19 (59.4%) as type 3B. Five (15.6 %) of the type 3B patients were assessed to have pelvic discontinuity.
Anteroposterior and lateral radiographs of the bilateral hips were taken for all patients immediately after the operation, as well as at 3, 6, and 12 months, and then annually thereafter, which is according to standard protocol at our hospital. The area surrounding the cup was divided into the 3 zones using the modified Delee-Charnley method, and zone 1 around the superior augments was further divided into two sub-zones 1A and 1B, with 1A in the superior lateral half and 1B in the inferior medial half of the bone-augment interface ( Fig. 1-3

Statistical Analysis
The horizontal and vertical COR location data were compared between the preoperative and postoperative measurements and also between the studied and the contralateral hips using the student t test. All statistical analyses were performed using the SPSS statistical software package (version 15.0; IBM, Armonk, NY, USA), and the level of significance was set at P < .05 for a single comparison.

Results
This cohort of patients was followed for an average of 40 months (range 25 -64 months) after surgery. All patients were able to ambulate freely at last follow up.  Table. 3). All components were found to be stable, with evidence of stable bone ingrowth identified for all patients based on the radiographic evaluation. There were 21 (65.6%) patients very satisfied with the outcome, 6 (18.8%) satisfied with the outcome, 2(6.3%) patients dissatisfied with the outcome. One patient was dissatisfied because of little improvement in hip function, and the other patient with rheumatoid arthritis and bilateral total knee arthroplasties was dissatisfied due to lower limb weakness. One  Table 3). There were 12 patients (37.5%) with regional radiolucent lines at the zone 1A augment region, and 1 (3.1%) patient found with regional radiolucent lines < 1mm wide in modified Delee-Charnley zone 2 of the acetabular cup, which was formed between the extruded cement and the corresponding radiopaque line of sclerotic bone. There were no radiolucent lines found at the zone 1B, zone 3, and the interface between cups and metal augments. All patients had sites of osteointegration between augments and bone, and between cups and host bone, identified on follow-up radiographies.
Between the EIEF and IIEF group, there was no significant difference in HSS scores, WOMAC scores, and satisfaction rate. For radiological results, there was no significant difference between the two groups in the horizontal and vertical location of COR, and orientation of cups (

Discussion
This study revealed that iliac extended fixation successfully reconstructed the hip COR, and achieved satisfactory clinical and radiographic outcomes for patients with severe superior acetabular bone defects in revision THA. There were no statistical differences in the major clinical and radiological results between the IIEF and EIEF technique groups, despite the latter group had more severe proximal migration of COR in preoperative radiographies. We did found statistical difference in the orientation of screws fixing the augments between IIEF and EIEF groups, which had potential biomechanical implications.
In this study, the concept of "iliac extended fixation" is proposed to differentiate it from junta-acetabular fixation, which is close to the original acetabulum. According Iliac extended fixation is defined as extending beyond 2 cm superior to the anatomic acetabular rim, based on anatomical and biomechanical rational [13][14][15][16][17][18][19][20].
Augments used at juxta-acetabular region can be considered a rim restoration procedure, while in iliac extended fixation, a superior or superior posterior/anterior supporting point is created by extending porous metal across the worst quality and quantity of bone potentially remote to acetabular rim. In comparison with juxtaacetabular fixation, iliac extended fixation requires more surgical exposure and is more difficult to obtain adequate primary mechanical stability. Our data revealed that bone ingrowth fixation was identified between the bone-cup and bone-augment interface in all patients, indicating that with iliac extended fixation achieved sufficient initial stability and subsequent biologic osteointegration without compromising COR and cup orientation ( Table 3-5).
The iliac extended fixation technique is aimed to reconstruct severe cranial acetabular defects while reconstructing the COR and cup orientation as optimal as we can. In the present study, the postoperative COR were significantly lower compared to the preoperative COR, more close to the anatomic COR ( Table 3). The hip biomechanics were therefore improved, which is consistent with previous reports [12]. Restoration of the COR helps decrease leg length discrepancy and avoid increasing the lever arm, which acts on the head/neck taper caused by  although not statistically significant (p = 0.075, Table. 5) It is still unknown whether the radiolucent lines in zone 1A will cause possible loosening of the cementless construct, which necessitates longer term of follow up.
This study had several limitations.

Ethics approval and consent to participate
The study was approved by the ethics committee of Beijing Jishuitan Hospital.
Written consent to participate forms were obtained from patients.

Consent for publication
Written consent form for publication forms were obtained from patients.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.