Pelvic Pseudotumor Associated With a Ceramic Bearing Total Hip

Abstract

Pseudotumors have been well documented to occur most frequently in metal-metal bearing total hip arthroplasties and less frequently in metal-polyethylene bearings. There are few cases in the literature of pseudotumors occurring in ceramic-ceramic articulations. We report a case of a large pelvic pseudotumor in a patient with a ceramic-ceramic bearing articulation in a 67-year-old man. In addition to the usual investigations, we did a detailed wear analysis of the ceramic implants and an examination of the soft tissues for particulate debris. The detailed wear analysis did show evidence of stripe wear; however, the volumetric wear was within the expected range. Synchrotron imaging identified strontium and zirconium debris arising from the ceramic surfaces. Although association does not mean causation, no other cause for the large pseudotumor could be identified and presumably represents an idiosyncratic reaction to ceramic debris.

Development of an adverse local tissue reaction (ALTR) is a notable and well-documented complication of total hip arthroplasties (THAs) using metal articulations although they have been also associated with metal-on-polyethylene (MoP) articulations. ALTRs are multifactorial and represent a spectrum of reactions that occur around the THA due to the inflammatory response and have inflammatory exudate and lymphocytes.^1,2 Although associated commonly with metal-on-metal (MoM) articulations, it is unclear whether the metal particles are themselves the cause of the lymphocytic reaction. ALTR cannot only cause the development of a soft-tissue mass that results in localized pain, but also damage to periprosthetic bone and soft tissue resulting in joint instability or total joint failure.^3,4

ALTRs are most common with MoM cobalt-chromium alloy articulations and less common with MoP implants. Literature suggests that ALTRs occur around MoP implants at the head-neck taper junction because of taper corrosion.^1,5,6 Because ALTRs occurred in relation to metal ion deposition in tissue surrounding the implants, ceramic-on-ceramic (CoC) articulations should have provided the solution we were looking for: an articular surface that would release inert wear debris.⁷

There are currently only four previous reported cases to our knowledge of an ALTR related to a CoC THA. Campbell et al⁸ reported the first case of an ALTR in the setting of a CoC THA without any metal ion release, but they failed to include any advanced wear analysis or analysis of the tissue sample for metal ions. Valcarenghi et al⁹ had the same pitfalls. Ishida et al¹⁰ reported a case of an ALTR after a CoC THA in the setting of femoral neck impingement likely leading to metal ion release. Movassaghi et al¹¹ reported a case of ALTR after a CoC THA in the setting of liner fragmentation leading to impingement, which likely resulted in metal ion release.

We present a well-documented case of an atypical ALTR in a patient with a CoC THA without any signs of metal ion release, including both an advanced wear analysis of the ceramic implants and of tissue microstructural and elemental analyses using synchrotron radiation techniques.

Case Report

Ethics review board approval was obtained for this investigation.

A 67-year-old otherwise healthy man had a left ceramic-ceramic bearing total hip arthroplasty at age 52 for osteoarthritis. Four years after the index procedure, he had a revision for “clunking” and squeaking. At the time of revision, the ceramic head was noted to have a large amount of scoring. Symptoms were thought related to subluxation because the head could be distracted from the acetabulum with traction. The femoral head was replaced with a 36-mm Biolox Delta with a plus five adapter sleeve. No mention was made of wear of the ceramic acetabular implant, which was not revised.

He was then asymptomatic for 9 years. He had a fall and developed back, hip, and leg pain. Investigations revealed a large amount of periacetabular osteolysis with a large intrapelvic mass pressing on the bladder. He was having symptoms of urinary retention.

He was referred to a cancer center for assessment. Investigations confirmed a large intrapelvic mass arising from the acetabulum. A needle biopsy found predominantly blood with fibrinous debris and hemosiderin and associated foreign body reaction. No evidence of malignancy or inflammatory pseudotumor was found. After this investigation, he was temporarily lost to follow-up in a cross-country move. He was referred for additional investigation and management.

His symptoms were less after the aspiration/needle biopsy, but were largely unchanged, reporting mainly of groin, buttock, and back pain. He required chronic narcotics to manage his pain. He was on tamsulosin and reported no additional urinary symptoms. Squeaking was not a main report, but did occur intermittently.

Physical Examination

The patient showed a Trendelenburg gait (long-standing). His flexion was 110°, internal rotation 20°, external rotation 30°, and abduction 25°. His C-reactive protein (high sensitivity) was 1.2 mg/L. Metal ions found were chromium < 1 ppb and cobalt < 1 ppb. Figure 1 shows AP pelvis radiograph demonstrating the total hip components to be in good position with no evidence of loosening. Acetabular osteolysis and large intrapelvic soft-tissue tumor are evident.

Figure 1.

AP pelvis radiograph. Note acetabular osteolysis and the large intrapelvic mass.

CT Scan

Figure 2 reveals a large cystic mass 11 × 8 × 11 cm³ causing a mass effect displacing the colon, bladder, and prostate, thought most likely to be a cystic ALTR.

Figure 2.

Image showing a CT scan. Note acetabular osteolysis and the large intrapelvic mass.

MRI

Figures 3 and 4 demonstrate erosion of the normal bone surrounding the medial margin of the acetabular implant with little remaining normal bone. A large thick-walled cyst with internal debris extends from the medial margin of the prosthesis through the acetabulum and then into the medial lower left pelvis. There is a second smaller cystic tissue in the region of the greater and lesser trochanters.

Figure 4.

Coronal MRI. The large intrapelvic mass is again demonstrated.

Figure 3.

Axial MRI. Note the large thick-walled intrapelvic mass with internal debris arising from the acetabulum displacing the bladder and rectum.

Revision Surgery

The implants were not loose. The acetabular implant was removed without complication. The femoral implant was not impinged on the acetabular implant. Extensive bone loss was noted about the acetabulum with a large medial defect. The consistency was mainly that of old hematoma and granulation tissue. This was bluntly scooped out. The medial defect was filled with sterile absorbable gelatin sponge and then morcellized allograft bone. A buttress augment was required to stabilize the noncemented trabecular metal hemispherical acetabular implant. The adapter sleeve was removed and showed the trunnion to be in good shape, save for a couple of scratches. The defects in the trochanters were curetted and packed with allograft chips. A 36-mm crosslinked polyethylene liner was placed in the acetabulum. A new adapter sleeve and 36-mm Biolox Delta ceramic head were used on the femoral side. Multiple intraoperative cultures for bacteria, fungi, and TB were negative. Bacterial cultures were held for 5 days and fungal cultures for 30 days.

Follow-up

Clinical

At the final follow-up 18 months after revision, he reported no pain. He was off all narcotics. He walked five miles daily with a cane, but still had a Trendelenburg gait. No evidence of recurrence of the ALTR was noted on repeat imaging. Unfortunately, he died of stroke 6 months later.

Light Microscopy

Two specimens were studied:

1. “Pseudotumor”: acellular fibrinous material with focal dystrophic microcalcifications. No wear debris were identified. No evidence of malignancy was noted.

2. Acetabular membrane: dense hypercellular scar tissue with aggregates of stromal hemosiderin, compatible with old hematoma. No wear debris were identified.

Wear Analysis

The acetabulum and femoral head were sent for detailed objective assessment of wear. Dimensional analysis was conducted using a coordinate measuring machine to measure volumetric wear and dimensional accuracies. Stripe wear was visible on the head and was measured to have contributed to a total material loss of 0.329 mm³. Wear appeared to have mainly developed about the wear scar/stripe wear on the articulating surface of the head (Figure 5). The articulating surface of the cup was measured to have suffered a total material loss of 3.421 mm³, which developed on the cup near the equator corresponding to the area of wear observed on the head. The total volumetric wear was considered minimal.¹² Overall, minimal volumetric loss was found, and bearing diameters were within tolerance.

Figure 5.

Image showing the wear pattern on the implant. Wear patch and stripe wear shown on the articulating surfaces of the revised ceramic cup and head (arrows).

Synchrotron Imaging

Tissue specimens were fixed in formaldehyde and prepared for synchrotron characterizations. For microstructural analysis, microcomputed tomography was performed at the BioMedical Imaging and Therapy Bending Magnet beamline.¹³ Samples were embedded in agar gel to immobilize them during the scanning. Incident radiograph beam was filtered through a 0.88-mm Al + 0.22 mm Mo foil (to prevent radiation damage on the tissue). Three thousand projections over 180° were collected from the samples with exposure time of 250 ms per projections (total scan time of ∼10 min). A white beam microscope (Optique Peter, France) coupled with a sCMOS PCO Edge 5.5 camera (PCO, Germany) and 50 µm LuAG:Ce scintillator at the magnification of 4.5x (effective pixel size of 1.44 µm) was used. The sample-to-detector distance was 5 cm.

Data reconstruction was done using UFO-KIT (Karlsruhe Institute of Technology, Germany).^14,15 Phase retrieval was used to enhance the contrast of the soft tissues. Image visualization and processing was performed in Avizo (version 2021.1, Thermo Fisher Scientific). High-density microparticles as small as 4 µm were found in the scar membrane tissue adjacent to the joint. No particles were found in the pseudotumor tissue (Figures 6 and7).

Figure 6.

Images showing synchrotron micro-computed tomography of the biopsies. Image A. and B. are from the membrane adjacent to the implant and clearly show (orange arrows) the ceramic debris as a result of the wear on the ball and socket implant. Image C. is from the pseudotumor and did not show any ceramic particles.

Figure 7.

3D images of microparticle distribution in membrane tissue. A, Image shows the maximum intensity projection in a membrane tissue adjacent to the implant (2D cross-section is shown in Figure 6A). B, Image only presents a 3D render of particles.

Synchrotron Spectroscopy

Radiograph fluorescence mapping and radiograph absorption spectroscopy measurements were performed on BioXAS-Main beamline. Incidence energy of the beam was fixed at 20 keV; size of the beam spot of the sample was limited to 2 x 0.5 mm (H x V). The sample was positioned at 45° relative to the incident beam; two 32-channel high purity germanium detectors were located on both sides of the sample at a distance of 85 mm.

One-dimensional fluorescence mapping shows distinctive Sr-Kα and Zr-Kα lines at the samples position (Figure 8). Presence of these lines in the background spectrum (outside of the sample) is caused by secondary scattering, similarly to bromine, known to be present in the sample holder material. X-ray absorption spectroscopy spectrum around Sr K edge was measured to evaluate strontium chemical speciation, which was determined to be an oxide or a Sr²⁺ salt.

Figure 8.

Graph showing x-ray fluorescence intensity at selected locations. Note the elevated levels of strontium and zirconium.

A hotspot of Sr on the x-ray fluorescence map is a clear indication that Sr in the sample exists as a relatively large particle or a group of particles, rather than being evenly distributed.

In addition to strontium and zirconium, the BioloxDelta ceramic head contains aluminum. No attempt was made to detect alumina in the sample because the synchrotron components contain alumina that make detecting small amounts in samples inaccurate.

Discussion

The development of ALTR have been well described in MoM bearing surfaces. A biologic, immunological response occurs with biomaterials and their wear products including polymers such as polyethylene and polymethyl methacrylate in addition to metallic and less commonly ceramic wear particles.^16,17 The term ALTR is used to describe the occurrence of abnormal periprosthetic soft-tissue lesions with or without a cystic element.¹⁸ Biopsy of these tissues are difficult to distinguish morphologically from a necrotic tumor, and thus, they were also called pseudotumors. Although there is no consensus on the exact process, they are characterized histologically by the presence of T-lymphocytes and macrophages with extensive necrosis, which would be in keeping with a metal-induced cytotoxic effect.^18,19 A proposed mechanism is that the metal ions that are generated as wear debris from the implant can induce mitochondrial stress that can generate a hypoxic-like condition in cells. This triggers the synthesis and secretion of cytokines that elicit inflammation in periprosthetic tissues.^11,20

The intuitive idea to combat metal ion debris is to use a non-metal articulation. However, there can be metal ion debris even when a non-MoM THA articulation is used. ALTRs have also been reported in ceramic-on-polyethylene articulations, but this was attributed to modular neck wear resulting in elevated metal ion levels.²¹ There are many reports of ALTRs in non-MoM THA articulations that identify trunnion wear as a source of metal ions.^5,22,23 Another case report proposes exposure and microabrasion of porous coating as a source for titanium metal debris as a source of ALTR.²⁴

Another proposed solution to the problem of inflammatory wear debris is typically to use an articulation whose wear debris is more inert such as ceramics. Classically, the advantages of CoC articulations include decreased friction and wear rate that result in a greatly decreased amount of wear debris and biological inertness that does not result in a granulomatous reaction.²⁵ However, newer research suggests that ceramic debris may not be as biologically inert as previously thought. Mochida et al¹⁶ found that ceramic debris still elicits a macrophage predominated response, which was confirmed by immunohistochemical staining. Petit et al²⁶ found that in vitro macrophage injection of ceramic particles increased proinflammatory cytokine production. The disadvantages of CoC include increased risk of fracture and greater occurrence of noises such as squeaking.²⁵

Campbell et al⁸ reported the first case of ALTR in a patient with a CoC THA. They concluded that ceramic wear debris must not be entirely inert because they determined that the ALTR occurred in the absence of a source for metal debris. Valcarenghi et al⁹ came to the same conclusion in their report of an ALTR in a patient with a CoC THA. It was a reasonable conclusion in the context of no obvious trunnion corrosion or particulate debris from impingement but cannot be completely excluded because advanced ceramic analysis and spectroscopy were not completed. Wear analysis of the ceramic implants in our case demonstrated minimal volumetric loss and no signs of impingement. Spectroscopic analysis of the pseudotumor showed the presence of strontium and zirconium particles, which are elements that make up the composition of ceramics material. These factors further support the conclusion that ceramic debris wear may not be inert and may even be capable of causing a proinflammatory state resulting in the development of an ALTR. The microparticles detected with synchrotron micro-computed tomography were sized as little as 4 µm; however, this should be emphasized that this is close to the spatial resolution limit of the detector combination used in this study and certainly there are much smaller particles, perhaps even in the nanometer range. Verifying that warrants repeating the imaging with a higher resolution detector.

Ishida et al¹⁰ reported a case of ALTR in a patient with a CoC THA likely due to metal particle release caused by impingement. The researchers included both histological analyses with metal staining and more advanced immunohistochemical staining. Using immunohistochemical staining with IL-17A proinflammatory cytokine, they were able to identify CD4⁺ T cells which correspond to hypersensitivity reactions to metal implants.²⁷ An improvement to our case would have been to include immunohistochemical staining in our workup as well, with the existing spectroscopy results to see whether there is a relationship between CD4⁺ T-cell predominance and the absence of metal ions in the ALTR tissue.

Movassaghi et al¹¹ reported the fourth case of an ALTR in a patient with a CoC THA but the first case in the setting of liner fragmentation and a prominent screw causing ceramic liner locking mechanism failure and backside wear. The researchers mentioned that the screw had minimal wear and that the ceramic implant had notable wear. It is hard to tell if debris from the screw, ceramic material, or both played a role in the development of the ALTR, if spectroscopic analysis of the ALTR tissue is not done.

Thus, ALTR etiologies can be due to many different causes such as trunnion corrosion, articulation impingement, articulate wear, porous coating exposure, screw prominence, and more recently hypothesized, ceramics material. Of the reported cases, ours gives more complete insight into the potential reactivity of ceramics material by excluding metal ion debris as a cause for ALTR for this patient. The case presented here is the first report that used spectroscopic analysis to most definitively say that no metal particles were in the ALTR tissue and that the only particles detected were strontium and zirconium which make up ceramic material. Additional studies are warranted to investigate the natural inflammatory response to ceramic debris.

Conclusion

Although association does not mean causation, no other cause for the large pseudotumor could be identified and presumably represents an idiosyncratic reaction to ceramic debris.