Aggressive granulomatous pseudotumor following total hip prosthesis: Diagnostic and therapeutic strategies from a single institution experience

Abstract

Introduction

Granuloma formation following total hip arthroplasty (THA) can occur regardless of the bearing surface. In very rare cases, extensive granulomas mimicking tumors may develop which are known as pseudotumors. The aim of this study is to report on these rare yet complex situations to stimulate reflection in diagnostic and therapeutic approaches.

Methods

A retrospective case series study of 5 patients (2 females, 3 males) of aggressive granulomatous lesions on primary prostheses and 4 revision prostheses for aseptic loosening, implanted at a mean age of 74.8 years (range 64–83). A clinical, functional, radiographic, and anatomopathological analysis were conducted.

Results

Four patients underwent revision surgery, while one died before intervention. Functional impairment was significant with a mean pre-reintervention Postel-Merle d'Aubigné (PMA) score of 8.2 (range 4–11). Femoral involvement was constant, associated with acetabular involvement in only one case. bony involvement was major, with a minimum of 5 Gruen zones affected, showing complete cortical lysis. Histopathological analysis consistently revealed multinucleated giant cells and macrophages, predominantly with metallic wear particles. In 2 cases, vascular involvement was suspected without a clearly delineated tumor. Two cases required femoral resection prostheses, one necessitated inter-ilio-abdominal amputation due to extensive soft tissue involvement, one underwent implant removal which failed to prevent progression, and one patient died before reoperation.

Conclusion

The volume of lesions in these 5 cases appears markedly larger than reported cases.

Extensive pseudo-tumoral granulomatous lesions following THA are rare but carry poor functional prognosis. Evaluation to rule out infection and neoplastic pathology is imperative. Surgical management often resembling tumor treatment should be considered. Early diagnosis is crucial to allow intervention before reaching the stage of massive prosthetic replacement.

Level of evidence

IV retrospective study.

1. Introduction

The biocompatibility of the materials used in total hip arthroplasties (THA) is still a topic of debate and research. Among the causes of mechanical failure of prostheses, aseptic loosening associated with peri-prosthetic granulomas has long been described.¹ It has been acknowledged that metallic wear debris or polyethylene (PE) is responsible for the granulomas. These particles create a local inflammation which, through the mediation of cytokines, leads to macrophagic activation and bone resorption.² Bone cement was also implicated before similar granuloma cases were reported with uncemented implants.^3,4 Thus, osteolysis is often multifactorial and influenced by the wear of the prosthesis.⁵

Granulomas typically manifest approximately five years following the implantation of THA, with a peak incidence observed at ten years..^6,7 These granulomas can achieve considerable size, encompassing the entirety of the proximal femoral metaphysis.⁸ This pathological condition, identified as aggressive granulomas, is histologically distinguished by the abnormal presence of giant multinucleated cells and macrophages.² The prevalence is estimated at 1 % of THA cases.⁷ Radiologically, they may mimic a tumoral lesion, as has been documented in prior studies.^1,9

These distinctive clinical presentations necessitate the exclusion of infectious and tumoral pathologies in the preoperative evaluation. Moreover, surgical management must be individualized and strategically planned to minimize the risk of surgical failure. We present a series of five cases of granulomas following THA, which mimicked tumors around the hip accompanied by the implemented diagnostic and surgical strategies. The objective of this study is to present these infrequent yet complex cases in order to foster thoughtful consideration in the diagnostic and therapeutic approaches.

2. Material and methods

This study was approved by our Institutional board (No.: 2024_060). We retrospectively analyzed cases of five patients with aggressive granulomatous lesions in the femur following THA, managed at our facility between January 2013 and November 2018.

Aggressive granulomatous lesions were defined as a pathological entity that can mimic a tumoral bone presentation, characterized by rapidly progressive extensive bone resorption at the proximal femur around the prosthesis, whether cemented or not, in the absence of infection.^2,3,10

There were 2 women and 3 men. The average age at the time of surgical revision was 72 years (62–83), and the average Body Mass Index (BMI) was 27.5 kg/m2 (17.3–35.6). The average follow-up period was 27.2 months (8-98). Regarding the implants used, there was one primary uncemented prosthesis and four prosthesis revisions, of which one was cemented.

We conducted a radiographic evaluation of the lytic zones according to the classification by Gruen et al.¹¹ and Paprosky¹² for the femur and the acetabulum.

We reported the preoperative and latest follow-up clinical scores using the Postel-Merle d'Aubigné (PMA) score¹³ the timing of osteolysis onset, and the types of prostheses. In all five cases, a histopathological analysis was conducted.

The indication for surgical revision was set for pain associated with aggressive granulomatosis with or without loosening. All procedures were performed by two senior surgeons. Continuous variables were expressed as means and ranges.

3. Results

All patients presented with hip pain as the primary clinical symptom, with a walking range of less than 500 m in three cases and less than 1 km in two cases. The average preoperative PMA score was 8.2 (4–11). All patients developed an aggressive granuloma around the hip resembling a tumoral lesion. The characteristics of these patients are detailed in Table 1. None of the patients exhibited signs of infection. The radiological aspect showed multifocal involvement in all cases. Preoperative radiographs are shown in Fig. 1, Fig. 2, and preoperative and perioperative data are presented in Table 2.

Table 1.

Demographic characteristics of the series.

case	Sex/Age	BMI	Initial diagnosis	Primary THA	Prosthesis (stem/acetabular cup) type	Ciment (stem/acetabular cup)	Bearing surface
A	M/62	35,6	osteonecrosis	yes	Uncemented stem/simple mobility	no/no	Metal/PE
B	F/70	28,3	hip dysplasia	no	Long revision stem/Mutars (Lumic)	no/yes	Metal/PE
C	M/80	30,8	coxarthrosis	no	Long stem/dual mobility	no/yes	Metal/PE
D	F/66	17,4	hip dysplasia	no	Long revision stem/cemented cup	no/yes	Metal/PE
E	M/83	25,6	coxarthrosis	no	Long revision stem/dual mobility	no/yes	Metal/PE

Fig. 1.

Preoperative radiological appearance of osteolyses at the upper end of the femur; case A; case B; case C; case D; case E.

Fig. 2.

CT Imaging of the osteolytic lesions; Case A; Case B; Case C; Case D; Case E.

Table 2.

Preoperative and perioperative data.

Case	Site of granuloma (Gruen zone)	Bone defect of femur regarding paprovsky classification	Bone defect of acetabulum regarding paprovsky classification	Preoperative PMA Score	Metallosis (yes/no)
A	1,6,7,8,14	2C	1	5	yes
B	1,4,7,8,9,10,11	3A	3B	10	yes
C	1,2,3,6,7,8,9	4	2A	11	yes
D	1,2,6,7,8,13,14	3B	1	11	yes
E	1,2,6,7,8,13,14	3B	2A	4	yes

Cases A and E underwent revision with a femoral resection prosthesis. Case D had an inter-ilio-abdominal amputation. Case B had their prosthesis removed without reimplantation, and Case C died before they could be treated. The radiological results are presented in Fig. 3.

Fig. 3.

Radiographs of cases A; case B; case D; case E.

For the two patients who underwent revision, the average follow-up was 14.5 months (8-21). Neither patient experienced a recurrence of the granulomatous lesion, and their average PMA score was 14.5 at the last follow-up. Surgical details and postoperative outcome are summarized in Table 3.

Table 3.

Surgical procedures and functional outcomes.

Case	Time Before Revision (months)	Lesion Volume (cm³)	Revision Technique	Revision Prosthesis (stem/cup)	Head Size (mm)	Bearing surface	Follow-up (months)	PMA Score	Complications
A	108 (9)	877	Bipolar revision	Massive prosthesis/Dual mobility	22	Metal/PE	24	14	no
B	96 (8)	2640	Removal	–	–	–	93	6	recurrence
C	96 (8)	2400	died due to Lymphoma	–	–	–	–	–	–
D	84 (7)	3230	Inter-ilio-abominal amputation	–	–	–	9	–	–
E	36 (3)	1881	Bipolar revision	Massive prosthesis/Dual mobility	22	Metal/PE	8	15	no

3.1. Details on the case series

In case A, the patient underwent THA in 2006 for osteonecrosis of the right hip. Pain appeared 7 years after the prosthesis was implanted, and both the pain and radiological osteolysis progressively worsened. In 2016, further imaging found no other lesion locations on bone scan, and hip aspiration culture was negative. Surgical biopsy revealed necrotic debris without histological evidence of tumor-type necrosis. We decided to perform a second surgical biopsy following embolization which identified a benign vascular tumor, either hemangioma or vascular malformation, associated with an organized hematoma. Immunohistochemical study analysis was positive for anti-CD31, and anti-ERG antibodies highlighted the character of hypervascularization. Anti-CD163 antibody highlighted the macrophagic elements within the fibrous tissue. In May 2017, after preoperative embolization of the lesion, the patient underwent a massive upper femur reconstruction prosthesis procedure with a cemented acetabular cup. The pathological analysis of the perioperative samples showed macrophagic granulomatous reaction related to metal debris, with no morphological evidence of infection or an immuno-allergic reaction.

In Case B, the patient underwent a THA in 1992 for a congenital hip dislocation. A bipolar revision was performed in 2006 due to aseptic loosening, followed by an acetabular revision in 2014 by LUMiC® acetabular implant (implantcast, Buxtehude, Germany) due to new acetabular loosening. The initial histopathological analysis identified a granulomatous reaction associated with metallic debris, without any signs of malignancy. Revision was complicated by recurrent dislocations and the progression of granulomatous osteolysis evident on radiographs, it was decided to remove the prosthesis in January 2016. The patient was lost to follow-up after 2016 and was recently hospitalized in February 2024 with a massive recurrence of a pseudotumor, necessitating further surgical management through excision (Fig. 4). Two weeks postoperatively, the patient underwent a new surgery for hematoma drainage, and bacteriological samples were collected, testing positive for Escherichia coli. A six-week course of antibiotic therapy was initiated, leading to a favorable outcome without the need for additional surgical intervention.

Fig. 4.

Case B: Coronal and axial CT of pseudotumor recurrence (A,B), peroperative macroscopic photo of the lesion (C,D).

In Case C, the patient underwent THA in 1985 for coxarthrosis. A bipolar revision was performed in 1998 due to aseptic loosening. The femoral component was changed in 2004 due to new loosening, and the acetabular cup was revised three months later with a dual mobility cup due to episodes of recurrent dislocation. A CT scan in 2011 revealed significant periprosthetic osteolysis. In 2012, a surgical biopsy was performed. Bacterial cultures were negative, and pathologic examination revealed a macrophagic granulomatous reaction to peri-prosthetic wear debris with hemorrhagic remodeling in the form of significant hemosiderin resorption. A unipolar acetabular revision was then performed for acetabular loosening. In 2015, the periprosthetic femoral osteolysis had progressed. A new biopsy showed hemosiderin deposits indicating bleeding and a macrophagic granulomatous reaction related to cement debris. There was no evidence suggesting an infection or tumor lesion. A revision surgery with an explant prosthesis was scheduled for 2017, but lymphoma was discovered preoperatively. The patient passed away from lymphoma before the surgery could be performed.

In Case D, the treatment was radical rather than conservative, involving an inter-ilio-abdominal amputation. The patient underwent THA in 1992, which was revised in 2006 due to a lytic lesion associated with loosening. Histopathological analysis at that time concluded it was a large, organized hematoma. By 2010, the lytic lesion continued to progress. A suspicion of a secondary tumoral lesion, appearing hypervascular on enhanced CT scans, could not be confirmed. A new surgical biopsy was decided upon after embolization of the lesion. The histopathological conclusions did not support a suspicion of malignancy but indicated an encapsulated hematoma and granulomatous reaction. Bacteriological and mycological tests were negative. Given the significant hemorrhagic risk of the lesion, its major extension to the soft tissues, and severe pain, it was decided not to pursue excision of the lesion combined with prosthetic reconstruction. The indication for an inter-ilio-abdominal disarticulation was set, and the procedure was performed in January 2013. The histological analysis of the lesion ultimately concluded an encapsulated hematoma with a granulomatous reaction typical of foreign body type from prosthetic wear, focally observed around the periphery of the prosthesis without signs of suspicious malignant tissue. Overall, the outcome has been satisfactory with no recurrence of the pseudotumoral lesion at the last follow-up.

In Case E, the patient underwent a THA in 1995, which was revised in 2015 due to loosening, using a long revision stem and a cemented dual mobility acetabulum and Kerboul acetabular cage. Postoperatively, the pain was not alleviated, and the patient was unable to resume walking. Following significant postoperative hip effusion, an aspiration was performed, but bacteriological and histopathological analyses were non-contributory. In November 2018, a unipolar change with a massive reconstruction prosthesis was carried out. During the procedure, a large necrotic mass was removed from the posterior compartment of the thigh adjacent to the osteolytic area. The histopathological analysis concluded that the material was fibrinous without any epithelial structures, atypical cells, inflammatory cellular elements, or any other atypical components.

4. Discussion

This study presents a series of five patients who experienced massive osteolysis around the hip after THA. The significance of this series lies in demonstrating that the management of this complication can be complex and may necessitate radical surgery.

The bony involvement was significant, always affecting at least five zones according to Gruen with complete cortical lysis. CT scan showed an extension of the granuloma into the soft tissues with an average volume of 2205 cm³ (877–3230). Several cases have been reported involving both cemented and uncemented implants.^3,8,10,14,15 In these series, the aggressive granulomas were radiologically less extensive than the five cases we report. The involvement of Gruen zones showed involvement to a maximum of four zones with complete lysis. None of the series reported granuloma affecting the soft tissues, suggesting that these might be more aggressive granulomas than those currently described, possibly even mimicking a tumoral lesion. Another explanation could be that diagnosis was made earlier in these series. In our series, the average time to surgical revision after the last prosthesis implantation was 7 years, comparable to published series where surgical revision time varied between 4.8 and 9.4 years. A third hypothesis might be that our series includes four cases of revision prostheses. The granulomatous lesions could have been pre-existing at the first revision but undiagnosed, potentially exacerbating the granulomatous process after the revision. The published studies report cases only following primary THA.

Eskola et al. defined this pathological entity as distinct from other periprosthetic osteolyses, both clinically, radiologically, and histopathologically.⁸ The histological and immunopathological characteristics of aggressive granulomas were described by Santavirta et al..² They identified connective tissue consisting of fibroblasts and reactive histiocytic mononuclear zones. Occasionally, small perivascular lymphocyte infiltrates were also present. In some respects, the aggressive granulomatous tissue resembled rheumatoid synovitis. Analyses revealed villous structures consisting of a layer of lining cells covering empty spaces in the highly vascularized granulomatous tissue. In our series, this latter characteristic was observed in Case D, where this hypervascularization precluded prosthetic revision due to the significant risk of major hemorrhage. In Case A, the second analysis identified a benign vascular tumor, with hypervascular features. This hypervascularization character could be present in a few cases and should be considered in the diagnostic approach to avoid significant operative difficulties that could lead to uncontrolled hemorrhage.

The origin of these aggressive osteolyses is multifactorial, and several etiologies have been suspected.⁵ Initially, cement was implicated in the first reported cases, linked to the fragmentation debris from the cement in aseptic loosening.¹ However, Polymethylmethacrylate appears relatively inert as long as the implants are not loosened, and several cases have been reported with a femoral stem that was not loosened at the time of revision.^10,16 Furthermore, cases of massive osteolysis have subsequently been reported with uncemented implants.^8,17

Wear debris is potentially a significant factor in the development of pseudotumoral reactions. Initially described in metal-on-metal bearing surface, the annual particle rate was reported to be 13.5 times higher than in metal-PE bearing surface.¹⁸ In our series, we exclusively had metal-PE bearing surface. The histopathological analyses conducted did not find PE debris but did identify metal debris in two patients and polymethylmethacrylate debris in one patient. These debris, through the macrophagic reaction they induce, have been reported by several authors as causing granulomatous reactions.^2,19,20

Thus, these pseudotumoral osteolyses can be provoked by a foreign body reaction to polymethylmethacrylate, PE, or metal from a THA.^14,21,22 PE debris induces a macrophagic response that releases prostaglandin E2, leading to bone resorption, which in turn causes loosening or osteointegration failure of the implant. Santavirta et al. demonstrated that the immune response does not depend on the chemical nature of the materials, observing a similar nonspecific immune response to both metal and polyethylene. However, they noted that, unlike aseptic loosening with less marked osteolysis, the analysis of aggressive granulomatous reactions revealed a significant presence of macrophages and multinucleated giant cells, with very little fibroblastic activity, which is nevertheless necessary for bone remodeling.^2,20

A hypersensitivity reaction to metal has been described as a potential cause of pseudotumoral osteolysis. There remains a question as to whether the immune response is an adaptive response to an excess of debris from high wear and is dose-dependent, or if it is an innate hypersensitivity response, which is independent of the extent of metal release and initiated at low levels of wear.⁵ This delayed type 4 hypersensitivity reaction has been described for metal-on-metal bearing prostheses.23, 24, 25 It appears to play a role in the genesis of pseudotumors.²⁶ Wilert et al. described a histological entity corresponding to this hypersensitivity reaction, termed ALVAL (aseptic lymphocyte-dominated vasculitis associated lesions), where a low rate of metal particles was detected.²³ Even though this immunological reaction seems to play a role in the genesis of pseudotumoral osteolysis, the fact that all our cases occurred in metal-PE bearing surface leads us to reconsider the immunopathological mechanism, which seems to be mixed and more related to the macrophagic reaction to wear debris.

Regarding patient-related factors, female gender has been identified as being at a higher risk for pseudotumoral osteolysis.²⁷ Nonetheless, the small sample size in our series is insufficient to substantiate this finding.

The primary complaint of patients is an increasing hip pain. Radiographic imaging often leaves little doubt about the diagnosis upon discovering massive osteolysis. Initially, it seems crucial to rule out infectious etiology. A detailed patient history of symptoms and a lab analysis including complete blood count to check for leukocytosis and C-reactive protein levels can facilitate initial screening. A biopsy can be performed for bacterial culture and histopathological analysis. The histological characteristics of an infected tissue differ from those of a tissue exhibiting an immuno-allergic or inflammatory reaction to foreign bodies. This biopsy can also help exclude primary or secondary tumoral lesions, especially if the patient has a history of cancer with common bone metastasis locations.

The diagnostic and therapeutic strategy remains challenging. In our series, one patient died before surgery could be performed. Another underwent resection arthroplasty for significant prosthetic instability and a deteriorated general condition, making a revision with a massive prosthesis too risky. One patient required disarticulation due to a large hypervascular lesion in the soft tissues, leading to non-conservative surgery via inter-ilio-abdominal amputation because of the major hemorrhagic risk. Two other cases involved using a massive prosthesis for the upper end of the femur, chosen due to extensive lysis and the need for an oncological resection of a granulomatous lesion to avoid progression. The cases reported in the literature were radiologically much less extensive and appeared earlier than those we report, which allowed for revision with a long femoral stem without the need for a massive tumor prosthesis, even though a case of resection arthroplasty was reported by Moholkar et al. due to a distal granuloma on a long stem.^3,7, 8, 9^,15

The main limitation of our study is the small sample size, although it involves a rare complication. Secondly, the histopathological analyses included immunohistological analysis in only one case, which does not allow us to compare our results with those of Santavirta et al.²

5. Conclusion

Extensive granulomatous pseudotumoral lesions following THA are rare complications with poor functional outcomes. A comprehensive evaluation to rule out infection and tumor pathology is required. Surgical management should be customized on a case-by-case basis. These patients must be diagnosed as early as possible to facilitate revision with a massive prosthesis and management in a specialized reference center.

Institutional review board statement

Ethical approval for this study was obtained from our institutional review board (n° 2024_060).

Informed consent statement

The authors certify that they have obtained all appropriate patients’ consent.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

CRediT authorship contribution statement

César Praz: data collection and analysis, Writing – review & editing, study design. Ramy Samargandi: Writing – review & editing, data collection and analysis. Philippe Rosset: study design, performing surgical interventions, manuscript proofreading and correction, Supervision, All authors have read and approved the final version. Louis-Romée Le Nail: study design, performing surgical interventions, manuscript proofreading and correction, Supervision.

Declaration of competing interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.