Early identification of patients at risk for impaired tendon healing and corresponding novel therapeutic approaches are urgent medical needs. This study aimed to clarify the role of CD3+ T-cells during acute Achilles tendon (AT) healing. Blood and hematoma aspirate were taken from 26 patients during AT reconstruction, and additional blood samples were obtained during clinical follow-up at 6, 26 and 52 weeks after surgery. T-cell subsets were analyzed by flow cytometry using CD3, CD4, CD8, CD11a, CD57 and CD28 antibodies. Clinical follow-up included functional tests, MRI assessments, and subjective questionnaires. In vitro, the functional behavior of patient-derived tenocytes was investigated in co-cultures with autologous unpolarized CD4+ or CD8+ T-cells, or IFNy-polarized CD8+ or IL17-polarized CD4+ Tcells (n=5-6). This included alterations in gene expression (qPCR), MMP secretion (ELISA), migration rate (scratch wound healing assay) or contractility (collagen gels). Analysis revealed that elevated CD4+ T-cell levels and reduced CD8+ T-cell levels (increased CD4/CD8 ratio) in hematoma aspirate and pre-operative blood were associated with inferior clinical outcomes regarding pain and function at 26 and 52 weeks. Increased levels of CD8+ -memory T-cell subpopulations in blood 6 weeks after surgery were associated with less tendon elongation. In vitro, tenocytes showed increased MMP1/2/3 levels and collagen III/I ratio in co-culture with unpolarized and/or IL17-polarized CD4+ T-cells compared to unpolarized CD8+ T-cells. This coincided with increased IL17 receptor expression in tenocytes co-cultured with CD4+ T-cells. Exposure of tenocytes to IL17-polarized CD4+ T-cells decreased their migration rate and increased their matrix contractility, especially compared to IFNy-polarized CD8+ T-cells. The CD4+ /CD8+ T-cell ratio could serve as prognostic marker for early identification of patients with impaired AT healing potential. Local reduction of CD4+ T-cell levels or their IL17 secretion represent a potential therapeutic approach to improve AT healing and to prevent weakening of the tendon ECM.

Tendon healing is a complex process that often results in compromised healing of the tendon tissue. It has recently been shown that temporal changes in the expression profile and the histological tissue quality of the tendons occur during the early healing process after acute Achilles tendon rupture. Whether these changes are accompanied by an altered healing process, is not yet known and was the aim of the present study.

Tendon biopsies were obtained from 24 patients with acute Achilles tendon rupture at the time of surgery (2–9 days after rupture) and examined histologically as well as on RNA level. Histologically, the tendon architecture, the amount of aligned collagen, glycosaminoglycan and fat as well as the cellularity, vascularity and immune cell infiltration were determined. On RNA level the expression of markers for the modeling/remodeling (MMPs and TIMPs), collagens (1, 3, 5), tendon markers (scleraxis, tenomodulin), pro- and anti-inflammatory markers (IL-1beta, IL6, IL10, IL33, TNFa, TGF-beta1, COX2) and immune cell markers (CD3, CD68, CD80, CD206) were analyzed by Real-Time PCR. To determine the clinical outcome, the patients were followed up 12 months after the operation and the following scores were recorded: Subjective score, Tegner score, Visual Analog Scale (VAS) pain, VAS function, Matles Test, Achilles tendon total rupture score (ATRS), Therman 100-points score, Heel rise test. Statistics: Spearman correlation analysis.

Correlation analysis shows that early post-rupture surgery is associated with better clinical outcome (ATRS Score: p=0.022). Histologically, a good functional healing outcome shows a positive correlation to the amount of aligned collagen (Heel Rise Test: p = 0.009) and glycosaminoglycans in the tendon (Heel Rise Test: p = 0.026, Matles difference: p = 0.029), as well as a negative correlation to the fat content (Thermann score: p = 0.018, subjective score: p = 0.027, VAS function: p = 0.031). On RNA level, a good healing outcome correlates with increased expression of MMP13, collagen 1, 3, 5 (Heel Rise Test: p = 0.019, p = 0.048, p = 0.030), and TIMP2 (Tegner Score: p = 0.040), TGF-beta1 (Thermann Score: p = 0.032) and CD80 (ATRS: p = 0.025, Thermann score:, p = 0.032). Whereas a limited healing outcome is associated with an increased expression of MMP2 (Heel Rise Test: p = 0.033), MMP3 (Matles Test: p=0.001, Heal Rise test p = 0.017), and IL33 (Tegner Score: p = 0.047).

The results of the study show a clear relationship between the tendon biology at the time of the surgery and the clinical and functional healing outcome 12 months after the operation. Especially matrix formation and remodeling play a crucial role, while the examined immunological factors seem to influence the tendon healing to a lesser extent. The modulation of matrix formation could potentially lead to improved treatment options in the future.

Despite the increasing availability of bone grafting materials, the regeneration of large bone defects remains a challenge. Especially infection prevention while fostering regeneration is a crucial issue. Therefore, loading of grafting material with antibiotics for direct delivery to the site of need is desired. This study evaluates the concept of local delivery using in vitro and in vivo investigations. We aim at verifying safety and reliability of a perioperative enrichment procedure of demineralized bone matrix (DBM) with gentamicin.

DBM (DBMputty, DIZG, Germany) was mixed with antibiotic using a syringe with an integrated mixing propeller (Medmix Systems, Switzerland). Gentamicin, as powder or solution, was mixed with DBM at different concentrations (25 −100 mg/g DBM), release and cytotoxicity was analyzed. For in vivo analysis, sterile drill hole defects (diameter: 6 mm, depth: 15 mm) were created in diaphyseal and metaphyseal bones of sheep (Pobloth et al. 2016). Defects (6 – 8 per group and time point) were filled with DBM or DBM enriched with gentamicin (50 mg/g DBM) or left untreated. After three and nine weeks, defect regeneration was analyzed by µCT and histology.

The release experiments revealed a burst release of gentamicin from DBM independent of the used amount, the sampling strategy, or the formulation (powder or solution). Gentamicin was almost completely released after three days in all set-ups. Eluates showed an antimicrobial activity against S. aureus over at least three days. Eluates had no negative effect on viability and alkaline phosphatase activity of osteoblast-like cells (partially published Bormann et al. 2014). µCT and histology of the drill hole defects revealed a reduced bone formation with gentamicin loaded DBM. After nine weeks significantly less mineralized tissue was detectable in metaphyseal defects of the gentamicin group. Histological evaluation revealed new bone formation starting at the edges of the drill holes and growing into the center over time. The amount of DBM decreased over time due to the active removal by osteoclasts while osteoblasts formed new bone.

Using this mixing procedure, loading of DBM was fast, reliable and possible during surgical setting. In vitro experiments revealed a burst and almost complete release after three days, antimicrobial activity and good biocompatibility of the eluates. Gentamicin/DBM concentration was in the range of clinically used antibiotic-loaded-cement for prophylaxis and treatment in joint replacement (Jiranek et al. 2006). The delayed healing seen in vivo was unexpected due to the good biocompatibility found in vitro. A reduced healing was also seen in spinal fusion where DBM was mixed with vancomycin (Shields et al. 2017), whereas DBM with gentamicin or DBM/bioactive glass with tobramycin had no negative effect on osteoinductivity or femur defect healing, respectively (Lewis et al. 2010, Shields et al. 2016). In conclusion, loading of DBM with gentamicin showed a proper antibiotic delivery over several days, covering the critical phase shortly after surgery. Due to the faster and complete release of the antibiotic compared to antibiotic loaded cement, the amount of antibiotic should be much lower in the DBM compared to cement.