Intractable neuropathic pain is a frequent consequence of nerve injury or disease. When peripheral nerves are injured, damaged axons undergo Wallerian degeneration. Schwann cells, mast cells, fibroblasts, keratinocytes and epithelial cells are activated leading to the generation of an "inflammatory soup" containing cytokines, chemokines and growth factors. These primary mediators sensitize sensory nerve endings, attract macrophages, neutrophils and lymphocytes, alter gene expression, promote post-translational modification of proteins, and alter ion channel function in primary afferent neurons. This leads to increased excitability and spontaneous activity and the generation of secondary mediators including colony stimulating factor 1 (CSF-1), chemokine C-C motif ligand 21 (CCL-21), Wnt3a, and Wnt5a. Release of these mediators from primary afferent neurons alters the properties of spinal microglial cells causing them to release tertiary mediators, in many situations via ATP-dependent mechanisms. Tertiary mediators such as BDNF, tumor necrosis factor α (TNF-α), interleukin 1β (IL-1β), and other Wnt ligands facilitate the generation and transmission of nociceptive information by increasing excitatory glutamatergic transmission and attenuating inhibitory GABA and glycinergic transmission in the spinal dorsal horn. This review focusses on activation of microglia by secondary mediators, release of tertiary mediators from microglia and a description of their actions in the spinal dorsal horn. Attention is drawn to the substantial differences in the precise roles of various mediators in males compared to females. At least 25 different mediators have been identified but the similarity of their actions at sensory nerve endings, in the dorsal root ganglia and in the spinal cord means there is considerable redundancy in the available mechanisms. Despite this, behavioral studies show that interruption of the actions of any single mediator can relieve signs of pain in experimental animals. We draw attention this paradox. It is difficult to explain how inactivation of one mediator can relieve pain when so many parallel pathways are available.