AIS patients from 13 centers were prospectively enrolled in a post-approval registry study to evaluate the continued safety and probable benefit of the MID-C system for five years postimplantation in AIS (NCT04296903). IRB approval and informed consent/assent were obtained as required. By the time of this report, 150/200 patients were enrolled in the study.

As per the study protocol, in case of reoperation, the surgeon may choose to obtain a biopsy based on their medical discretion upon visualization of the tissue. This report includes an analysis of all tissue samples collected by the time of this report (N = 7). Each sample was accompanied by a form completed by the treating surgeon, describing in detail the surgical site at the time of implant removal and the surrounding tissue.

Tissue samples were collected during the reoperation procedure, fixed in 10% neutral buffered formalin, and transferred to AnaPath Services GmbH, Switzerland, for analysis. Each sample was trimmed at four approximately equidistant locations (levels 1–4), dehydrated, embedded in paraffin wax, sectioned at an approximate thickness of 2–4 µm, and stained with hematoxylin and eosin (HE). In addition, Masson’s trichrome (MT) stain was used for fibrosis scoring. The resulting slides were quality-checked under the microscope and then dispatched to the study pathologist for examination under light microscopy.

The histologically stained microscope slides were scanned (whole-slide imaged, WSI) with an Olympus Sideview VS200 slide scanner using a VS-264C camera and the 20× objective.

Tissue analysis was performed according to a prospective evaluation protocol, GLP, and relevant sections of ISO10993-6:2016, Biological evaluation of medical devices—Part 6: Tests for local effects after implantation. The histology sections were evaluated independently by two board-certified pathologists.

The tissue origin of the samples was determined and grouped into the following categories: fibrous connective tissue, loose connective tissue, skeletal muscle, adipose tissue, nerves, and bone fragments.

Host reaction fibrosis was differentiated from physiologic fibrous connective tissue (ligaments, parts of the zygapophyseal joint capsule that represent real joints as so-called plane joints) by its morphology. Physiologic fibrous connective tissue has few vessels, is cell poor, and is highly organized. Fibrosis contrarily is composed of unorganized, plump fibroblasts with newly formed vessels. Physiologic fibrous connective tissue was not scored for the purpose of tissue response as it is autochthonous tissue.

Artifacts were present in all samples and consisted of tissue basophilia and coagulation at the sample border induced by monopolar electrocautery or, in a few cases, drying artifacts (tissue basophilia without coagulation) (Weber et al. manuscript in preparation and [7]).

A semi-quantitative histopathology evaluation was performed according to an adapted ISO 10993-6 scoring system (Table 1), to score the host inflammatory response. Additional parameters were included in the evaluation, specifically granuloma, mineralization, hemorrhage, fibrin, fatty infiltrate, bone fragments, edema, and hemosiderin.

Besides the host response, the presence, distribution, and size of wear particles were scored and included in the evaluation. The score for each tissue sample was the average score over all four levels.

EDX evaluation was performed on tissue samples of four of the seven evaluated patients (i.e., patient IDs: 021-A001, 091-A002, 090-A004, and 091-A007) if wear particles were present on the histological tissue slides.

For these patients, one additional unstained and non-cover slipped tissue section for EDX analysis was prepared for the level with the highest “Wear particles—presence” score (if several levels had the same highest score, the level used for EDX evaluation was selected by the study pathologist).

Tissue sections were silver sputter-coated at an approximate thickness of 5 nm using a plasma sputter coater (PLASMATOOL-SC, Ingbuero Peter Liebscher). The tissue sections were then inserted into the scanning electron microscope (SEM) (Phenom Pharos Desktop SEM with FEG Source, Thermo Fisher Scientific) for energy dispersive X-ray (EDX) analysis operated by FEI Phenom ProSuite (2.9.0, Thermo Fisher Scientific) and Phenom Element Identification (3.8.6.0, Thermo Fisher Scientific) software.

Regions of interest identical to those containing the particles identified in the WSI optical scans were allocated in the SEM viewer by matching the allocation and morphology of the WSI. Per sample, a ten-point element analysis of the visible particles was conducted in back-scattered electron (BSE) mode, at 60 Pa vacuum and 15 keV beam.