To the editor: Coates and colleagues report four kidney graft recipients with post-transplantation lower limbs transient joint pain. Bone scintigraphy and magnetic resonance imaging (MRI) of the symptomatic joints showed increased uptake of the tracer and bone marrow edema, respectively. The authors therefore advise MRI in kidney recipients with lower limb pain to distinguish transient bone marrow edema from osteonecrosis (1). We already made this suggestion almost 10 years ago (2, 3). A syndrome of lower limbs osteoarticular pain appears in up to 10% of kidney recipients shortly after transplantation: clinical examination is often unremarkable, and paraclinic investigations show (a) epiphyseal patchy osteoporosis at X-ray, (b) increased focal or diffuse uptake of the tracer at bone scintigraphy, and (c) areas of low-signal intensity at T1-weighted images (with a corresponding high-intensity T2-weighted signal) at MRI, in 41, 81 and 83% of symptomatic joints, respectively (2, 3). Magnetic resonance imaging abnormalities are consistent with areas of medullary edema and/or hemorrhage. Within these areas some bands of low-signal intensity can be seen. These bands presumably correspond to epiphyseal impactions, the so-called ‘microfracture’ (2, 3). Clinical recovery usually occurs within several weeks; periosteal reactions can be seen at X-ray whereas bone scintigraphy and MRI completely normalize (2, 3). Can cyclosporine be the cause of this syndrome? Coates et al. mention that symptoms resolved with cyclosporine dose adjustment. Cyclosporine has previously been incriminated: its prescription coincided in similar cases with onset of the symptoms, it was the sole immunosuppressant used, or symptoms resolved in some patients concomitantly with a decrease in cyclosporine dosage. The occurrence of the same syndrome before the cyclosporine era or to patients given tacrolimus (4) probably refutes this hypothesis. If it is not cyclosporine, what is the cause of this syndrome? Because of the MRI findings we initially proposed a microtraumatic hypothesis (2, 3). The demonstration in some patients of very subtle bands of hyposignal within the bone marrow edema areas further supports the notion of impaction. In this context, the recent findings of Yamamoto are of interest (5). He correlated the MRI and histopathologic features of a subchondral femoral head insufficiency fracture (in a nontransplanted patient): exactly as in our patients, MRI showed bone marrow infiltration with a low-intensity signal on T1-weighted images, which histologically corresponds to callus, edema, and vascular granulation into viable bone and marrow tissue. Even more interesting, some of the T1-weighted images disclosed a focal band beneath the articular cartilage, which correspond to a fracture line and its associated repair tissue (5). We now recommend our patients to refrain from excessive physical activity shortly after transplantation to minimize risk of locomotor injury. This is important, because in a minority of patients the evolution of this transient lower limb joint pain syndrome is not as benign, and epiphyseal osteonecrosis may occur (2, 4).