Objectives The pathophysiology of radiculopathy associated with lumbar spinal stenosis and lumbar disc herniation is incompletely understood. The goal of the present study was to establish a chronic spinal nerve root compression model that can mimic lumbar disc herniation or spinal stenosis using silicone tube compression. We also try to link the pathology changes of damaged nerve root with the reaction of microglia in spinal cord in same rat at different time points. Methods Thirty rats were used in this study. The L5 nerve roots (dorsal and ventral) were exposed by hemilaminectomy; the diameter of the L5 nerve root was measured at the 2 mm proximal from the dorsal root ganglia. The dorsal and ventral nerve roots of L5 were compressed using a silicone tube, and the sham group was only exposed dorsal and ventral roots of L5. Five rats from the sham group were perfused at 8 days after surgery, and 25 rats from the model groups were perfused at 3, 8, 12, 45 days, and 5 months after surgery, each model group was composed of 5 rats according to the time point. The L5 spinal cord segments and nerve root that compressed by silicone tube were harvested from the same rat. Microglia and neuron in the spinal cord were stained by immunohistochemistry, and the nerve root was shown by electron microscope. Results In sham-operated rat, the arrangement of axon and myelin sheath is normal, the ventral root is mainly composed of large axon ([6 lm) and it is composed of 46.3 % of all the axons of the ventral root; the average myelin thickness of large axon is 1.86 lm; the dorsal root is mainly composed of medium (2-3.9 or 4-5.9 lm) axons and they are composed of 79.1 % of all the axons of the dorsal root; the average myelin thickness of this category is 0.94 or 1.55 lm. The average myelin thickness of large axon in ventral root reduced to 0.97 and 1.19 lm from more than 1.86 lm after compression for 3 and 8 days separately. Most of myelin sheath disappeared after 12 days of compression; the myelin sheath was partly restored at 45 days after compression which the myelin sheath thickness of large axons in ventral root was 0.47 lm. The medium category in dorsal root reduced to 0.59 or 0.72 lm from 0.94 lm, and 1.55 lm after compression for 3 days (p \ 0.05 to p \ 0.0001). The medium category axon in dorsal root is also 0.47 lm after compression for 45 days (p B 0.0001). The myelin sheath was almost totally restored at the 5 months of compression; the myelin sheath thickness returned to normal and the axons were intact in structure under EM. The number of Iba1-positive microglia increased by 18.69, 40.44, and 18.49 % after compression for 3, 8, and 12 days separately in the ipsilateral dorsal horn and 21.26, 32.15, 22.87 % in ventral horns, and the activation of microglia was also prominent in contralateral sides of the dorsal and ventral horn at