The aim of this study was to present the current status of algorithms for the treatment of PJI in certified arthroplasty centers. All German Centers certified by EndoCert have been included. These 515 centers are representing the most of the high-volume arthroplasty centers regarding the THR’s and TKR’s including the respective revisions (1, 15). All EndoCert approved centers are obliged to send their implant data of primary and revision procedures to the German joint register EPRD. When comparing the register data with the knowledge of center-specific algorithms, these results may offer successful approaches for the treatment of PJI in the future. This is one of the few and one of the most comprehensive national studies on treatment of PJI (17-19).
The response rate of the survey questionnaires was 100%. Only question two was not fully evaluable with 90.3% and did not reach the maximum overall level; all other questions were answered with a 100% evaluable response rate.
In the registers, such as the Swedish Hip and Swedish Knee Arthroplasty Register (SHAR and SKAR (20, 21)), the American Joint Replacement Registry (AJRR (22)), the Australian Orthopedic Association National Joint Replacement Registry (AOANJRR (23)), as well as in the German register (EPRD (24)), the term revision is not consistently defined. There is no definition if a revision is reported as implant-related or if it is considered infected. Infections are recorded as the cause of revisions but are not further discussed in detail. The change of inlays in the case of infections does not count as revision in the registers and is not further evaluated with respect to success or failure of the procedure; soft tissue surgery is not recorded in some cases either (20). Only the SKAR addresses this problem in a separate section (21). Revision procedures and algorithms are not recorded in any of the registers (20 – 24).
Due to the lack of uniform therapy algorithms for the treatment of PJI, the selected procedure is often based on traditions, experience and preferences of the surgeon and the institution (25, 26). While standards in primary joint replacement must already be established at the time of the initial certification of the center, the goal of consistent, reproducible and clinically coordinated diagnostic and therapeutic processes in treatment of PJI has not yet been sufficiently achieved. There is a development towards standardization in the field of septic endoprosthetic cases, at least in the certified centers. The high number of centers (n = 115) without a defined therapy concept clearly illustrates the need of compulsory and approved standards. Within the complication management the EndoCert-centers had to define an intraclinical treatment concept which is reviewed within the scope of regular audits. On an international and national level, this issue is of great importance (27) and a standardized approach is necessary when treating PJI.
This study showed that almost all EndoCert centers (97.6%) use DAIR in early PJI. In addition to joint debridement, the replacement of modular prosthetic components is an essential part of the concept (98%) in order to eradicate a potential source of infection and enable more extensive debridement (29). A large number of centers stated that they would perform this procedure within 6 weeks after initial implantation while retaining the implant components that are well fixed in the bone. The literature states that after 6 weeks, the chance of success of implant-preserving surgery is reduced to 40% (30). This is in agreement with the findings of implant-associated biofilm research showing that biofilm forms on the biomaterial surfaces early after implantation and is particularly advanced after 4 weeks (31, 32). In centers with more than 200 revisions per year, partial replacement is used up to a maximum of 4 weeks, or in an interval of 4-10 weeks after primary implantation. The use of DAIR in cases of late hematogenous endoprosthesis infections, which can impress like an early infection with a fulminant course, was not considered in this survey, but represents a further treatment strategy.
In case that DAIR is not effective, all centers change their strategy to one- or two-stage revisions of all implant parts. The significance of both procedures in PJI is intensely discussed; some studies describe similar or higher success rates for one-stage revision (33 - 35). Patient selection for eligibility of the procedure precedes the one-stage exchange. This is crucial for the success of the therapy (36 - 38). The present study has shown that patient selection in the case of a one-stage revision is also of great significance for the EndoCert centers. The definition of early and late infection is not uniformly applied (28, 29). There is no clear statement on a duration regarded as early infection by the centers and the corresponding period to perform the one-stage exchange. In order to be able to determine uniform therapy algorithms, it is crucial to determine a consistent definition of early infection. The one-stage exchange is only conducted at half of the EPZs for single cases and is not regarded as a 100% equivalent alternative to the two-stage procedure.
The two-stage exchange is implemented in 100% of the EPZ in Germany. This procedure can be applied to a wider range of patients and leads more reliably to successful treatment of the infection, especially in cases with difficult-to-treat bacteria (39, 40). Shorter OR time, the possibility of fractional expansion, management of the infection situation, the systematic administration of appropriate antibiotics both intravenously and locally with antibiotic-reinforced spacers, and an interval to reduce the risk of infection persistence prior to reinstallation result in an increased use in the EPZ. In cases of severe infections, poor soft tissue conditions or (partially) resistant bacteria, the literature reports a clear advantage of a two-stage procedure (41). A disadvantage is the repeated anesthesia and relatively limited mobility in the (spacer) interval. The duration of the implant-free interval is 4 to 8 weeks in 61% of the centers and corresponds to the reports in the literature (25, 39, 42, 43). The advantage of the two-stage revision is the shorter duration of the individual procedures and the possibility of obtaining several delayed tissue samples from different areas of the wound region, resulting in early and better detection of infection and appropriate administration of antibiotics (39).
The duration of the implant-free interval showed a high discrepancy of the indicated durations of 4 days to 120 days; a further differentiation of the values could not be determined from the questionnaire. On average, the spacer was in place for 42 days before the next revision, corresponding to the time recommended in the literature (39, 42).
PMMA spacers containing antibiotics are used during the interim interval in more than 90% of the centers and the Girdlestone situation is also implemented in 23% of the centers as alternative solution. If PMMA is used, the hospitals applied individually shaped cement spacers (61%) and pre-formed spacers (42% of the centers). When inserted in the knee and hip, this spacer should increase mobility in the interim phase (44). However, the cast spacer can only be used for minor bone defects, while individual spacers are adapted to the specific defects and can bridge larger distances and ensure the release of antibiotics in the overall defect (45). It is discussed whether the movement of the spacer has favorable properties for the healing of the infection (46, 47).
In addition to surgical decontamination, the management of suitable antibiotics plays an important role in the treatment of infections. The duration of therapy as stated in the literature varies from 14 days to 3 months (25, 39, 42, 48). A distinction between removal and replacement is advisable and is carried out by 94% of the centers. 280 hospitals (45.6%) choose an individual application of antibiotics. At the time of the survey, the rate of standardization was still very low, which could partly be due to the various literature. The duration of antibiotic treatment after removal of the prosthesis is largely homogeneous; 85% of the centers surveyed state a duration of 3 to 6 weeks and only 5% state a longer duration of treatment. In contrast, the centers are much more inconsistent after reimplantation. The majority (39.9%) stated a duration of antibiotics of up to 2 weeks whereas 37.8% stated a duration of 4 to 6 weeks. Further prospective comparing studies are necessary to determine the best outcome of the treatment concepts. As the duration of the course of antibiotics increases, the antibiotic-associated rate of adverse side effects that pose a risk to the patient increases (49).