Introduction The application of multipotent mesenchymal progenitor cells is among the most promising options for treatment of tendinopathy. Studies in large animal models and equine patients suffering from natural tendinopathy suggest that these cells promote tendon regeneration and reduce re-injury rates [e.g. Crovace, 2010; Godwin, 2012]. However, the fate of the applied cells and their mechanism of action are not yet fully understood.

Magnetic resonance imaging (MRI) is a standard modality for monitoring tendon healing. Moreover, MRI can be used for non-invasive cell tracking over several weeks when cells are labelled with superparamagnetic iron (SPIO) particles. However, to our knowlegde, no studies on MRI cell tracking following tendinopathy treatment have been published to date, potentially because visualisation of SPIO-labelled cells in tendons is complicated by the hypointense MRI signal of healthy tendon tissue. We aimed to overcome this difficulty by applying an imaging technique using the “magic angle effect” and to perform cell tracking and parallel monitoring of tendon healing by MRI in the equine model for the first time.

Methods Initially, tendon explants were seeded with different numbers of SPIO-labelled cells in vitro and subjected to MRI and histology (n = 3). MRI was performed using low- and high-field magnets (0.27 T, 3 T and 7 T). Images were obtained in different sequences and with the tendons positioned at 90° and 54° angles to the magnetic field and cell traceability was assessed and validated.

Thereafter, labelled progenitor cells were applied locally for treatment of tendon disease in 3 horses with induced tendinopathy and further 3 horses suffering from natural tendinopathy. Control tendons were injected with non-labelled cells or serum. Clinical, ultrasonographical and low-field MRI assessment was performed regularly. MRI was performed directly before and after cell application and at 2, 4, 8 and 12 weeks using T1-, T2*- and T2-weighted (w) as well as STIR sequences.

Results The in vitro study demonstrated that cells are traceable in tendon tissue by low- and high-field MRI. When tendons were positioned at 54°, the magic angle effect lead to hyperintense signal from the tendon tissue, allowing to distinguish the hypointense artefacts generated by the labelled cells from the surrounding tendon. Quantitative image analysis showed good correlations between seeded cell numbers, the extent of hypointense artefacts in MRI and the extent of iron staining in histology. Still, sensitivity of cell detection strongly depended on the sequences and the field strength of the magnet used. In 0.27 T low-field MRI, which was also used for the in vivo study, T2*-w sequences obtained at 54° were most sensitive and allowed detection of 10⁶ and 10⁵ cells reliably, while the signal obtained from 10⁴ cells was weak.

The in vivo study confirmed that the established MRI technique is feasible for cell tracking in the living animal. Hypointense artefacts related to the applied cells could be detected at the injection site directly after injection and remained visible during the follow-up period of 12 weeks in T2*-weighted images. In 2 animals, part of the injected cells also appeared to migrate towards damaged structures in the surrounding tissue. Migration of cells to more distant locations was not evident. Furthermore, good progress in tendon regeneration could be observed in T2-w and STIR images, matching the clinical and ultrasonographical findings.

Discussion The results indicate that the cells generally remain at the injection site within the damaged tendon tissue but are also capable of migration and homing to structures in the surrounding tissue. Further studies including histology will need to confirm that the hypointense artefacts in MR images correspond to the injected cells at all timepoints.

References Crovace et al. Vet Med Int. 2010;2010:250978

Godwin et al. Equine Vet J. 2012;44:25–32