1 Introduction

Advances in the diagnosis and treatment of cancer have led to increased survival rates and improved quality of life for patients. Bone metastasis surgery has two primary objectives: the resection of the tumor to improve survival in selected cases and the reduction or elimination of pain and instability caused by the disease.

The pelvis is the second most common site of bone metastasis after the spine.^1,9,16 Periacetabular metastases (PAMs) are frequently associated with severe pain and impaired ambulatory function, as the pelvis is a weight-bearing joint complex predisposed to pathological fractures,² resulting in impaired mechanical stability of the pelvic ring. The management of PAMs is complex and challenging, requiring a multidisciplinary approach in specialized referral units. Before deciding on surgical treatment, the patient's estimated survival and the size of the lesion should be considered. The median survival for pelvic pathological fractures is typically less than two years.³

The conventional Harrington classification⁴ (Fig. 1) is the most commonly used system for classifying these lesions. It differentiates four classes based on the affected region and surgical strategy. Class I lesions involve cavitary defects, and the recommended surgical approach is intralesional excision followed by conventional cemented total hip prosthesis. Class II lesions affect the medial wall and quadrilateral plate but spare the lateral wall or acetabular roof. For these lesions, excision and total hip prosthesis with an anti-protrusion device, such as a flanged cup, are recommended. Class III lesions, which involve both the medial and lateral walls and the acetabular roof, are the most challenging to reconstruct. Intralesional excision, total hip prosthesis, and acetabuloplasty with large Steinmann pins are suggested for these cases. Class IV metastases are solitary lesions that require en-bloc resection and may be candidates for reconstruction with a saddle prosthesis.

Close

Fig. 1 Author's illustration about the Harrington's expanded classification of periacetabular metastases. The examples of affected areas and lesions are shown according to each classification type. In the expanded classification, we propose adding the following categories: A) joint involvement; B) involvement of Enneking zone 1; C) presence of pathological fracture at diagnosis; and X) none of the above. THA, Total hip arthroplasty.

Export to PPT

The Harrington technique,⁴ described in 1981, was initially developed to reconstruct acetabular defects using Steinmann pins threaded with cement as an augmentation of a total hip prosthesis, allowing load transmission to healthy pelvic bone. However, the original technique is now rarely used, and several modifications have been proposed by different authors^5–8 to optimize mechanical stability, tailored to each specific case.

This study aims to describe the outcomes of our series of patients with PAMs and hematologic tumors and to propose an extension of the conventional Harrington classification.

2 Materials and methods

2.2

2.2 Surgical procedure

Small acetabular lesions that do not compromise load-bearing areas can be treated non-surgically. However, larger lesions that compromise stability require surgical reconstruction. Periacetabular metastases were graded according to the Harrington classification, with four additional categories.-Category A: joint involvement;-Category B: Enneking zone 1 involvement;-Category C: presence of pathological acetabular fracture at diagnosis:-Category X: none of the above.

The appropriate Harrington class and category were determined based on radiological evidence, primarily from plain radiography and CT scans. Various surgical options have been described for metastatic disease of the acetabulum, ranging from isolated percutaneous cementoplasty to surgeries associated with high comorbidity and complication rates, such as the modified Harrington technique or endoprosthesis implantation.

Pain relief and restoration of ambulation are the main goals of reconstructive procedures.⁹ In this study, we employed the following techniques: isolated cementoplasty, the modified Harrington technique, total hip prosthesis with a reinforcement ring, total hip prosthesis with a reinforcement ring combined with the modified Harrington technique, and reverse cone prosthesis CONED (Fig. 2). Preoperative embolization was necessary for some patients. All procedures were performed by a team of orthopedic surgeons specialized in oncology and orthoplastic management. Most patients received postoperative radiotherapy. Follow-up visits were conducted routinely at 2, 3, 6, and 12 months, and annually thereafter. We used the Katagiri score, a key prognostic tool for metastatic bone disease, and ASA score to assess the patient's overall health.

Close

Fig. 2 Radiographs of reconstructive procedures. A: Supraacetabular cementoplasty. B: Modified Harrington technique using Steinmann nails threaded with cement. C: Reinforced ring hip prosthesis. D: Combined modified Harrington technique and reinforcement ring hip prosthesis. E: Reversed CONED prosthesis.

Export to PPT

3 Results

Of the 28 patients with PAMs (Table 1), 25.00 % of cases were secondary to breast cancer, 17.85 % to gastrointestinal cancer, 14.28 % to multiple myeloma, 10.71 % to renal cancer, 7.14 % to lung cancer, 7.14 % to plasmacytoma, 3.57 % to cervical cancer, 3.57 % to prostate cancer, 3.57 % to adrenal gland cancer, 3.57 % to osteosarcoma, and in one case, the primary tumor could not be identified. The primary tumor could not be identified in a 48-year-old female patient (no. 23 in Table 2), who passed away 4 months after diagnosis. Despite ongoing investigations, the origin of the primary tumor remained unknown. Multiple myeloma and plasmacytoma cases were included due to their similar surgical approach to metastatic bone disease.

3.5

3.5 Survival

Survival rates were 64.30 % at 12 months, 57.10 % at 24 months, and 28.62 % at five years. The eight patients who survived more than five years had a mean Katagiri score of 3.12 ± 1.45 (range: 1−5) compared to the overall mean of 3.92 ± 2.05 (range: 1−8). In this group of eight patients, four patients were treated using CONED prostheses, three cementoplasty procedures, and one modified Harrington technique. The postoperative Kaplan-Meier survival curve for patients with PAMs showed a median survival of 24 months (Fig. 3). The median survival for males was eight months, and 48 months for females, though this was not statistically significant (log-rank p = 0.97, Gehan-Breslow 0.06) (Fig. 4).

Close

Fig. 3 Kaplan-Meier survival curves by category of the proposed expanded Harrington classification. A: Joint involvement. B: involvement of Enneking zone 1. C: presence of pathological fracture at diagnosis. D: All categories combined.

Export to PPT

Close

Fig. 4 Kaplan-Meier survival curves by reconstruction method. Number 1 (blue) is Combined modified Harrington technique and reinforcement device. Number 2 (red) is total hip prosthesis with a reinforcement ring. Number 3 (light green) is CONED prosthesis. Number 4 (lilac) is supraacetabular cementoplasty. Number 5 (dark green) is modified Harrington technique.

Export to PPT

The median survival for patients with specific primary conditions was as follows.-Primary breast cancer, 47 ± 45 months;-Primary digestive cancer, 33 ± 62 months;-Multiple myeloma, 34 ± 25 months;-Primary renal cancer, 22 ± 33 months.

Three patients (10.71 %) died before the third postoperative month. Two had primary renal cancer, and one had primary digestive cancer. All three had an ASA score of IV. The mean Katagiri score for these patients was 6.33 ± 2.88 (range: 3−8) compared to the overall mean of 3.92 ± 2.05 (range: 1−8). Complications negatively impacted patient survival, with a mean survival of 3 ± 1 months compared to 43 ± 45 months for the entire sample with a p value < 0.01 in the Mann-Whitney U test.

The median survival of patients with PAMs by category was as follows (Fig. 3) (Table 2).-Category A (n = 5), 24 ± 22 months;-Category B (n = 5), 36 ± 31 months;-Category C (n = 7), 33 ± 52 months;-Category X (n = 11), 69 ± 63 months.

None of the added categories were associated with survival, as no significant differences were found. Similarly, no statistically significant differences in survival were observed according to the type of reconstruction used. Sex-based differences were also not observed, as the Breslow (Generalized Wilcoxon) test yielded a chi-square value of 3.598 (p = 0.058). The median survival was 8 months for men and 48 months for women, with no statistically significant difference (Log Rank p = 0.97, Breslow p = 0.058). In the subgroup analysis of patients over 60, a significant association was found between Category A and survival (chi-square p = 0.026). Significant differences were identified with respect to survival for the Katagiri score (p < 0.001) and the ASA score (p = 0.014). No significant differences were observed between the newly created variables and survival at 6 months, 12 months, 2, 3, 5, and 10 years.

4 Discussion

Among the most relevant findings of our study are the variety of surgical reconstruction techniques, the long-term follow-up, the absence of implant or systemic infections, and the rate of ambulation recovery, which compares favorably with other studies on periacetabular metastases (PAMs) (Table 3). The majority of patients with PAMs present with involvement of Enneking zones II and III at the time of diagnosis, due to their anatomical proximity. Tumors involving zone I pose an additional mechanical challenge and complicate reconstruction, as they affect the iliac bone.

Unlike most published studies, we employed a wide variety of reconstruction techniques, including isolated cementoplasty, modified Harrington technique, total hip prosthesis with a reinforcement ring, the combination of the two previous, and CONED prosthesis. In contrast, authors such as Lavignac et al.,¹⁰ Wegrzyn et al.,¹¹ and Tsagonis et al.,¹² among others,^13–15 primarily used total hip arthroplasty for reconstruction. However, Shahid et al.¹⁶ demonstrated good functional outcomes, including pain reduction in 92.85 % of cases and an ambulation rate of 96.40 %, in a series where cementoplasty, resection and disarticulation, in addition to total hip arthroplasty, were used.

The therapeutic management of PAMs includes three approaches: palliative, minimally invasive, or major reconstructive techniques. Employing a variety of surgical techniques allows for more individualized treatment, tailoring the reconstruction to the specific needs of each patient. Minimally invasive methods, such as radiofrequency ablation, cryoablation, embolization, and cementoplasty, have shown good short-term outcomes by improving pain and providing immediate stability.^17–20 However, these techniques do not offer local control of the disease. Cementoplasty, either alone or with augmentation screws,²¹ is a simple and effective procedure for small lesions with a low complication rate. Some authors use augmentation balloons, similar to those used in kyphoplasty, to optimize cementation.²²

The modified Harrington technique has evolved, with the introduction of dual mobility systems, reinforcement rings, and other anti-protrusion systems reducing the rates of prosthetic dislocation and aseptic loosening.^11,12 Porous tantalum implants offer good biological fixation due to their friction surface and low modulus of elasticity and can be used for large bone defects or as a salvage option after the failure of other techniques.^23,24 While porous tantalum implants are associated with a high rate of complications, they are particularly useful in advanced lesions. In our series, three cases of implant stability failure were successfully treated with CONED prosthesis. Shahid¹⁶ reported a case of CONED prosthesis dislocation within the first postoperative month, attributed to acetabular component anteversion.

Several authors advocate for surgical treatment strategies to avoid undertreatment or overtreatment in PAMs.^9,25–27 While robotic-assisted reconstructions have recently been published,²⁸ more scientific evidence is needed to integrate this approach into a treatment algorithm.

The mean follow-up of 43 months in our series differs from other studies that report means of 33,¹¹ 28,¹⁰ or 12 months.^12,16 In addition to the ASA score, we used the Katagiri score, a key prognostic tool for metastatic bone disease, to assess the patient's overall health. Despite advances in cancer treatment, patients with PAMs and hematological tumors continue to have poor prognoses, as reflected in the modest improvement in median survival published in recent studies.^10–13 Published survival rates two years after surgery range from 30.00 % to 55.00 %.^7,16,29,30

The absence of implant or systemic infection in our series is noteworthy. The rate of medical and mechanical complications, as well as mortality in the first three postoperative months, aligns with the rates reported in the literature. Our ambulation rate of 82.14 % is lower than those reported in larger series,^10–13 which could be attributed to the fact that two of the five non-ambulatory patients experienced medical complications. Additionally, all five had preoperative ASA score of IV and died before the fourth postoperative month.

Our proposed expanded Harrington classification aims to standardize the approach for the international scientific community. This expanded classification includes four categories that help guide reconstruction choices and predict survival outcomes. Based on radiological evaluations, such as plain radiography and CT, the classification correlates with prognosis. Category A indicates the worst prognosis (mean survival of 24 months), while Category X suggests the best prognosis (mean survival of 69 months) (Fig. 3).

In our series, the most common reconstruction method was a total hip prosthesis with a reinforcement ring. Our objective is to standardize the reconstruction approach based on the categories added to Harrington's classification. Cementoplasty would be an appropriate method for patients without joint involvement, pathological fractures, or Enneking zone 1 involvement—corresponding to category X. Conventional arthroplasty could be utilized for cases with joint involvement but adequate bone reserve, classified as category A. Patients with severe zone 1 involvement and insufficient bone stock (category B) would be candidates for Girdlestone-type resection arthroplasty or, in very select cases, custom prostheses. CONED prostheses would be indicated in cases with significant zone 2 involvement and/or pathological fractures with joint destruction, corresponding to category C.

The conventional Harrington classification can be limited and sometimes confusing. Some authors²⁷ have proposed treatment algorithms where the Harrington classification is relegated to a less relevant level or even suggest class-specific modifications.³⁰ Wei et al. proposed differentiating Harrington class III lesions into two subgroups: Class IIIa, where the lesion is distal to the inferior border of the sacroiliac joint, and Class IIIb, which corresponds to a lesion extending proximal to the inferior border of the sacroiliac joint. According to this author, Class IIIa lesions are candidates for modified Harrington technique reconstruction, while en-bloc resection and endoprosthetic reconstruction are the best treatment approaches for Class IIIb lesions. We consider this modified classification incomplete because it does not address the remaining classes or other reconstructive procedures, and it considers a p-value <0.10 as significant.

Some of the limitations of our study include those inherent to the retrospective single-center design, small sample size, lack of control group, diagnostic heterogeneity, low frequency of the pathology and the length of the study period, which may introduce bias in surgical indications and overlook technological advances. Additionally, comparing the mechanical stability of different reconstructive procedures proved challenging. A larger sample would improve the analysis of subcategories, provide greater statistical power, and validate the proposed extension of the Harrington Classification.

5 Conclusion

In this study, the majority of patients were able to maintain ambulation, with the choice of reconstruction largely determined by the extent of the bony defect. Taking into account the patient's overall condition and prognosis, we propose an extension of the conventional Harrington classification by adding four categories. These additional categories, based on radiological findings, provide guidance for selecting the appropriate reconstruction method and offer insight into the potential survival outcomes for patients.

Patient's consent

Informed consent was obtained from all the patients of the study.

Authorship

All authors had full access to the data in the study and took responsibility for the integrity of the data and the accuracy of the data analysis. Conceptualization, JGP and JLCP; Methodology, JGP, RGM and JLCP; Investigation, JGP, JC and SL; Formal Analysis, JGP, JC and JLCP; Resources, JLCP; Writing - Original Draft, JGP and RGM; Writing - Review & Editing, JGP, RGM, JLCP and AGL; Visualization, JLCP; Supervision, JLCP and AGL; Funding Acquisition, AGL.

Ethical Statement

Hereby, I Javier Gutierrez-Pereira consciously assure that for the manuscript Periacetabular metastases: proposed extension of the Harrington Classification the following is fulfilled.1)This material is the authors' own original work, which has not been previously published elsewhere.2)The paper is not currently being considered for publication elsewhere.3)The paper reflects the authors' own research and analysis in a truthful and complete manner.4)The paper properly credits the meaningful contributions of co-authors and co-researchers.5)The results are appropriately placed in the context of prior and existing research.6)All sources used are properly disclosed (correct citation). Literally copying of text must be indicated as such by using quotation marks and giving proper reference.7)All authors have been personally and actively involved in substantial work leading to the paper, and will take public responsibility for its content.

The violation of the Ethical Statement rules may result in severe consequences. To verify originality, your article may be checked by the originality detection software iThenticate. See also http://www.elsevier.com/editors/plagdetect. I agree with the above statements and declare that this submission follows the policies of Solid State Ionics as outlined in the Guide for Authors and in the Ethical Statement.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.