Plexus injury results in lifelong suffering from flaccid paralysis, sensory loss, and intractable pain. For this clinical problem, regenerative medicine concepts set high expectations. However, it is completely unknown how dorsal root ganglia (DRG) are affected by accidental deafferentation. Here, we investigated human DRG in a clinically characterized cohort of patients with plexus injury. Avulsed DRG from 13 patients were collected during reconstructive nerve surgery. DRG were analyzed using large-scale microscopy, deep learning-based bioimage analysis, and RNA fragment sequencing of histopathological slices. In about half of the patients, we found a complete loss of DRG units consisting of neurons, satellite glial cells (SGC), and macrophages. The DRG cells were replaced by mesodermal/connective tissue. In the other half of the patients, the cellular units were well preserved and we found no gliosis and no significant neuronal loss. Furthermore, the expression of subtype-specific sensory neuron marker genes was preserved. However, downregulation of neuronal attributes associated with ion transport, synapses, and neuronal projection indicated a dormant neuronal phenotype. Injured DRG showed signs of ongoing inflammation and connective tissue remodeling. Patients with ‘neuronal preservation’ had less pain than patients with ‘neuronal loss’. Arm function improved after the nerve reconstruction, but the pain phenotype did not. With this study, we call for early intervention after injury to protect the DRG from complete cell loss. Pain patients may benefit from anti-inflammatory therapy. Future regenerative medicine concepts will need at least two translational directions: reafferentation of existing DRG units or replacement of the entire DRG.