Chronic subarachnoid midazolam (Dormicum) in the rat. Morphologic evidence of spinal cord neurotoxicity

In humans, the benzodiazepine midazolam has been reported to exert an antinociceptive action after subarachnoid injections. It has been shown that subarachnoid midazolam given to rabbits produces significant pathology in spinal cord morphology, as detected with light microscopy. In order to further characterize these changes, this study was performed, using a more sensitive histologic technique, including electron microscopy as well as unbiased morphometry.

The histopathology of the rat lumbar spinal cord was investigated after chronic subarachnoid administration of a commercially available preparation of midazolam. After daily injections of 100 micrograms of midazolam, the animals were transcardially perfused on the twentieth day with a mixture of formaldehyde and glutaraldehyde.

Morphometric evaluation of cell number and mean cell volume (MCV) by the disector method revealed a significant lower (P < .05) cell number and a tendency toward higher MCV in the midazolam-injected group (n = 6), compared to the rats injected with saline (n = 6). The higher MCV, in combination with a reduced number of nerve cells, indicated a loss of small neurons. The electron microscopic findings confirmed that midazolam caused neuronal death, since degenerated cell somata, fibers, and terminals were observed in most of the rats. Furthermore, an increased number of microglial cells phagocytosing nerve structures were also seen mainly in the dorsal horn.

The authors found that chronic subarachnoid administration of midazolam gives objective signs of neurotoxicity in the rat spinal cord. The authors' findings are in contrast to those of an earlier light microscopic study in the rat. The present results emphasize both the necessity of morphometric and ultrastructural studies before spinal administration of novel drugs to humans and the neurotoxic potential of midazolam. Since neurotoxicity of midazolam now has been demonstrated in both rats and rabbits, there may be reason to be sceptical of the use of subarachnoid midazolam in humans.