Characteristics and Epidemiology of Megaprostheses Infections: A Systematic Review

Background: Megaprostheses were first employed in oncological orthopedic surgery, but more recently, additional applications have arisen. These implants are not without any risks and device failure is quite frequent. The most feared complication is undoubtedly the implants’ infection; however, the exact incidence is still unknown. This systematic review aims to estimate in the current literature the overall incidence of megaprosthesis infections and to investigate possible risk/protective factors. Methods: We conducted a systematic search for studies published from July 1971 to December 2023 using specific keywords. To be included, studies needed to report either the megaprosthesis anatomical site, and/or whether the megaprosthesis was coated, and/or the surgical indication as oncological or non-oncological reasons. Results: The initial literature search resulted in 1281 studies. We evaluated 10,456 patients and the overall infection rate was 12%. In cancer patients, the infection rate was 22%, while in non-oncological patients, this was 16% (trauma 12%, mechanical failure 17%, prosthetic joint infections 26%). The overall infection rates comparing coated and uncoated implants were 10% and 12.5%, respectively. Conclusions: The number of megaprosthesis implants is increasing considerably. In traumatological patients, the infection rate is lower compared to all the other subgroups, while the infection rate remains higher in the cancer patient group. As these devices become more common, focused studies exploring epidemiological data, clinical outcomes, and long-term complications are needed to address the uncertainties in prevention and management.


Introduction
Massive long bone defects pose significant challenges for reconstruction in the orthopedic field.Various techniques and strategies have been adopted to treat these bone defects, such as autograft and allogeneic bone grafting, bone transport, and the use of standard prostheses and megaprostheses (MP) [1].Megaprostheses, also referred to as tumor endoprostheses, are systems that allow special segmental bone and joint replacement, which were initially developed for lower limb and then upper limb salvage.First employed in oncological orthopedic surgery in the 1960s [2], megaprostheses, along with the advent of adjuvant therapies, have dramatically changed the management of bone tumors, which previously condemned patients to limb amputation.Since the 1990s, megaprostheses have become the gold standard for reconstruction after the resection of primary malignant bone tumors.Occasionally, further oncological indications have come into place, such as reconstruction after the eradication of locally aggressive benign bone tumors, malignant soft tissue tumors, or bone metastasis [3,4].More recently, additional applications have arisen for megaprostheses implants as a last resort for revision arthroplasty in selected cases, where extensive bone loss or poor bone quality jeopardizes the success of conventional joint arthroprosthesis, or for trauma/revision trauma surgery with severe soft tissue damage, severely comminuted fractures, or similar bone quality concerns [5,6].As expected, however, the implant of a megaprostheses is not without any risks, and device failure is more frequent than in other primary joint arthroplasties [7,8].Recently, Henderson et al. [9] proposed a classification system for megaprostheses failure.The latter can be divided into mechanical causes, such as soft tissue failure, aseptic loosening, and structural failure, and nonmechanical, such as infection and tumor progression.Mechanical challenges such as dislocation are quite common: the rate of hip dislocation after proximal femur replacement ranges from 6% to 42% [10,11].Another major challenge lies in the recovery of good joint function, especially when the knee is involved.Scar tissue, joint problems from previous deformities or concomitant degeneration, and muscle loss with loss of contractile function are common in patients undergoing megaprostheses placement and have a deep impact in joint mobility recovery.Failure can also originate from the low quality of superficial soft tissue coverage, which can be compromised following trauma, septic conditions, or recurrent surgeries [12], or by adjuvant therapy in oncologic diseases, such as radiotherapy [13].When the implant is placed for tumors, recurrence and progression are other possible causes of implant failure and reduced survival [9].In addition, one of the most feared complications is megaprostheses infection, which is even more frequent than periprosthetic infections according to multiple studies.However, the exact incidence of periprosthetic infection in patients with megaprostheses is still unknown and may range from 3% [14] to higher than 30% [9,15].Infections are facilitated by a longer surgery time and wider soft tissue dissection compared to other orthopedic procedures, along with a usually more immunocompromised host who may need multiple revision surgeries [16].Although indications for the implantation of megaprostheses are increasing, this type of prostheses is still quite rare.Therefore, the exact epidemiology and incidence rate of megaprostheses infection are largely unknown due to a lack of literature.To date, it is also not yet known whether there are specific risk factors that may negatively influence the risk of infection for these implants.This systematic narrative review first and foremost aims to estimate in the current literature the overall incidence of megaprostheses infections.Furthermore, it aims to investigate possible risk/protective factors with respect to the onset of infections such as the impact of the anatomical location, the indications for placement, and the presence/absence of implant coating.Greater clarity in the matter could help to establish preventive measures and to give clues in terms of management and outcomes.

Materials and Methods
We conducted a comprehensive systematic search on four databases (PubMed, Embase, Scopus, and Web of Science) using the search keywords ((megaprostheses) OR (megaprosthesis) OR (resection arthroplasty) AND (infection)).No restrictions were applied to the publication dates.Therefore, the research included studies published from July 1971 until December 2023.The bibliographies of the selected studies were manually searched to identify additional papers not found through the ordinary search.The titles of the journal, authors' names, and supporting institutions were known throughout the whole process.This systematic review was conducted according to the latest version of Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) [17], as reported in Figure 1.The 27-item PRISMA checklist of this systematic review can be found in the Supplementary Materials.This study was not registered in any protocol database; therefore, there is no registration number.Inclusion criteria were specified as follows: studies published in Italian, English, or French, describing the number of infected implants out of the total number of megaprostheses implanted.Moreover, studies had to report either the megaprostheses anatomical site (total femur, proximal tibia, distal femur, proximal femur, proximal humerus, custom megaprosthesis after hemipelvectomy, and other), and/or whether the megaprostheses were coated, and/or the surgical indication as oncological or non-oncological reasons.Exclusion criteria were established as follows: review articles (systematic or narrative), meta-analysis, letters, case reports, notes, conference papers, editorials, and conference abstracts; cadaveric and animal studies; full text not available; and finally, articles published in any language other than the previously stated ones.The primary outcome is to determine the overall incidence of infection per implanted megaprostheses.The secondary outcome is to establish the occurrence of megaprosthesis infection per implanted megaprostheses categorized by anatomy (total femur, proximal tibia, distal femur, proximal femur, proximal humerus, custom megaprostheses after hemipelvectomy), coating (presence or absence), and reason for implantation (oncological and non-oncological, with the latter further divided into trauma, periprosthetic joint infections (PJIs), and mechanical failure).Abstracts and full texts were independently screened by three authors (L.C., F.S., and G.M.), and any arising conflict was solved by consensus with a fourth author (F.T.).All the selected studies were retrospectively analyzed by two authors (M.B.B. and R.V.) who then extracted and entered the data in an Excel worksheet.The collected data included authors' list, year of publication, and the number of infected prostheses out of the total number of megaprostheses implanted.In addition, the implanted and then infected megaprostheses were specifically studied according to the anatomical district involved, the presence or absence of coating of the megaprostheses, and indication for megaprostheses implantation.Lastly, the data sheet was reviewed by three authors (A.Z., M.F., and G.M.) who agreed and validated the extracted data.
studies published in Italian, English, or French, describing the number of infected implants out of the total number of megaprostheses implanted.Moreover, studies had to report either the megaprostheses anatomical site (total femur, proximal tibia, distal femur, proximal femur, proximal humerus, custom megaprosthesis after hemipelvectomy, and other), and/or whether the megaprostheses were coated, and/or the surgical indication as oncological or non-oncological reasons.Exclusion criteria were established as follows: review articles (systematic or narrative), meta-analysis, letters, case reports, notes, conference papers, editorials, and conference abstracts; cadaveric and animal studies; full text not available; and finally, articles published in any language other than the previously stated ones.The primary outcome is to determine the overall incidence of infection per implanted megaprostheses.The secondary outcome is to establish the occurrence of megaprosthesis infection per implanted megaprostheses categorized by anatomy (total femur, proximal tibia, distal femur, proximal femur, proximal humerus, custom megaprostheses after hemipelvectomy), coating (presence or absence), and reason for implantation (oncological and non-oncological, with the latter further divided into trauma, periprosthetic joint infections (PJIs), and mechanical failure).Abstracts and full texts were independently screened by three authors (L.C., F.S., and G.M.), and any arising conflict was solved by consensus with a fourth author (F.T.).All the selected studies were retrospectively analyzed by two authors (M.B.B. and R.V.) who then extracted and entered the data in an Excel worksheet.The collected data included authors' list, year of publication, and the number of infected prostheses out of the total number of megaprostheses implanted.In addition, the implanted and then infected megaprostheses were specifically studied according to the anatomical district involved, the presence or absence of coating of the megaprostheses, and indication for megaprostheses implantation.Lastly, the data sheet was reviewed by three authors (A.Z., M.F., and G.M.) who agreed and validated the extracted data.

Search and Literature Selection
The initial literature search resulted in 1281 studies.After reviewing all the abstracts, the restricted research retrieved 959 studies.Once duplicates were removed and the articles were screened for inclusion and exclusion criteria, 138 studies remained, and full texts were assessed for eligibility (Figure 1).Finally, a total of 91 articles were included in this systematic review [15,.

Study Characteristics
We evaluated 10,456 patients who had undergone limb reconstruction with modular megaprosthesis.The infection rate was 12% (1277/10,456).Our results are summarized in Table 1.

The Initial Diagnosis
Most patients had an indication to implant a modular megaprosthesis after tumor resection (4283/10,456, 41%) including either primary tumors or metastatic disease.In cancer patients, the infection rate was 22% (941/4283).Among all patients who instead had a megaprosthesis implanted in a non-oncological setting, the infection rate was 16% (221/1377).More specifically, 652 modular megaprostheses were implanted for significant bone loss due to major trauma, failed osteosynthesis with a nonunion, or even periprosthetic fractures.The infection rate in this group of patients was 12% (76/652).Furthermore, 142 megaprostheses were implanted in the case of mechanical failure, and among them, 24 later became infected with an infection rate of 17%.Finally, 318 modular implants were used in periprosthetic joint infections (PJIs) and the re-infection rate was found to be 26% (84/318).

Anatomical Classification
Among all the modular megaprostheses included in the study, 7766 were implanted in the lower limbs and 265 in the upper limbs.The overall infection rates were 15% (1155/7766) and 10% (27/265), respectively.Among the lower limb implants, 139 were total femur replacements of which 39 became infected (28%), 5780 were either proximal tibia or distal femur replacements with 885 ending up being infected (15%), and finally, 1402 were proximal femur replacements with an infection rate of 13% (187/1402).Among the 142 megaprostheses of proximal humerus included, 13 became infected (9%).Hemipelvectomy and reconstruction with a custom megaprosthesis was performed in 21 patients and among them 8 became infected (38%).

Discussion
Megaprosthesis replacement was originally used for reconstructive treatment in limb salvage surgery following soft tissue and bone tumor resections [108].To date, indications are becoming broader and megaprostheses are now used in the management of bone loss such as in severe trauma, periprosthetic fracture, and arthroplasty revision, or in the case of periprosthetic joint infection [109].However, the patient selection in this type of surgery is crucial because a possible revision surgery could be technically challenging with a high clinical burden and low functional recovery, especially in the frail population [110,111].Given its increasing frequency, orthopedic surgeons should be aware of the risk of infection of a megaprosthesis as this is a complication that can lead to adverse clinical and functional outcomes, and in some cases, even to the patient's death [112].
As far as we know, no recent epidemiological reviews in the literature have fully addressed this problem.We recruited 91 articles for a total of 10,456 patients with different features of indications, anatomy, and coating.Among these, the overall megaprosthesis infection rate was 12%.
Our data showed different infection rates based on the different initial diagnosis.
Our study showed that there exists an important difference in terms of the infection rate between cancer patients (22%) and non-cancer patients (16%).The poor general clinical conditions of an oncological patient may explain such data.A recent review by Gonzalez MR et al. identified several modifiable and unmodifiable risk factors such as chemotherapy, radiation therapy, immunosuppression and sufficient soft tissue coverage, operative time, and length of stay, which have to be considered when a cancer patient has to undergo surgery [113].Previous studies indicated an incidence of infections of approximately 8% in megaprostheses implanted for oncological reasons, far different from our data [9].Certainly, the increased use of these prostheses even only in the oncological context together with the increased survival of these patients leads to an increase in complications and thus also infections.On the other hand, Vayhsa et al. instead recruited studies focusing on the implantation of lower limb megaprostheses in non-cancer patients finding an infection rate of 18.5%, comparable to our findings [114].We can therefore conclude that cancer patients with megaprostheses have a higher risk of implant infections compared to non-oncological patients.Concerning more specifically the non-oncological group, the infection rate in patients with megaprostheses implanted for traumatological indication is 12%, which is higher than expected.Indeed, recent findings by Sambri et al. [109] showed instead an incidence of 8.2% that is explainable, however, by a difference in the sample.Considering mechanical failure as an indication for megaprosthesis implants, the infection incidence rate was 17%.Studies on the same indication in the lower limb described a similar rate (18%) [49,113], while few data exist concerning the upper limb, describing very few cases of infection [114,115].In the subgroup with PJI as the lead indication for megaprosthesis implants, the infection rate was 26%; higher than the other two subgroups, as expected, although in line with the existing literature [20].
Furthermore, our data showed different infection rates based on the anatomical site.Lower limb megaprostheses demonstrated an infection rate of 15% (1155/7766 patients), compared to 10% (27/265) in the upper limb.Similar values were found by Schmidt-Braekling et al. who showed a lower limb infection rate higher than 17% [116].However, the data in the literature are not unique.Indeed, Windagher et al. reported a rate of infection in patients treated with a megaprosthesis after a distal femur periprosthetic fracture that ranged from 6.6% after 1 year to 45% after a mean follow-up of 34 months [117].
Going into further detail, we found a higher rate of infection in knee megaprostheses compared to hip megaprostheses.These data can be explained both by the greater number of knee megaprostheses implanted and also because the knee is a most insidious anatomical site, which often does not provide adequate soft tissue coverage [118].
Finally, considering the outcomes for coated or uncoated megaprostheses, we found a higher infection rate for uncoated megaprostheses.Unfortunately, to date there are no univocal results in the literature on this matter.Not all studies included specified whether the megaprostheses were coated or uncoated; however, we assumed that when unspecified, the implants were uncoated.A total of 9049 patients had uncoated megaprostheses implanted with an infection rate of 12.5% (1135/9049 patients).The infection rate appears to decrease for coated implants such that in this group the infection rate was 10% (141/1407 patients).Lex JR et al. performed a review on the different outcomes between coated or uncoated megaprostheses, investigating major anti-bacterial coatings currently in use, including 11 studies: only 2 studies demonstrated better outcomes in coated megaprostheses compared to uncoated ones [115].Furthermore, Fiore M et al. did not find any statistically significant difference between the outcomes in coated megaprostheses compared to uncoated ones, concluding that coated megaprostheses should only be used in selected cases [119].No clear conclusions could be drawn on the difference of outcomes between coated and uncoated; however, our findings seem to suggest a potential protective role of coated prostheses as already mentioned by other authors.
Our paper has several limitations.This is a systematic review that was not registered.Furthermore, the articles' quality has not been assessed.Moreover, although it was possible to extrapolate the number of implanted megaprostheses and the rate of infections in all included works, unfortunately, it was not always possible, for instance, to trace the reasons why the prosthesis was implanted.Therefore, such detailed tables were purposely included.In addition, due to the wide heterogeneity of the articles included in this review, it was not possible to assess the follow-up and then relate all our data to it.This is mostly because too often the follow-up was not mentioned or, unfortunately, was not relatable to the individual groups.
The present literature review focuses on the significantly increasing number of megaprostheses implantations and, accordingly, the increasing number of implant failures due to infections.To the best of our knowledge, there is no review in the literature that includes more than 10,000 patients undergoing megaprosthesis replacement.However, the large number of works examined, and the wide variability in the examined data and outcomes do not allow us to draw clear conclusions.In the future, the opportunity to identify not only predisposing but also protective factors with respect to megaprostheses will make it possible to develop standardized treatment algorithms to reduce the risks for patients in general, and specifically the infectious one.These would enable us to select patients and thus associate each one with a specific device and surgical approach to optimize treatment and outcomes.

Conclusions
In conclusion, the number of megaprosthesis implants is increasing considerably.While in the past megaprosthesis implants were exclusive to cancer patients, to date, they are increasing in number by increasing the indications.Nowadays, megaprostheses are mostly used for lower limb replacements, especially around the knee.Megaprostheses implanted as a consequence of severe trauma are increasing significantly, second only to cancer patients.In traumatological patients, the infection rate is lower than in both cancer patients and patients with PJI but also in patients with the mechanical failure of prostheses.Knee megaprostheses have a higher infection risk than hip megaprostheses, which can help the surgeon to decide on different perioperative management approaches for the hip and the knee.Coated implants have a slightly lower infection rate than uncoated implants and should be considered in patients at high risk of infection.As these devices become more common, focused studies exploring epidemiological data, clinical outcomes, and long-term complications are needed to address the uncertainties in prevention and management.

Figure 1 .
Figure 1.Search and literature selection.

Figure 1 .
Figure 1.Search and literature selection.

Table 1 .
Infection rate in relation to principal diagnosis, anatomy, and implant characteristics.