In 1960, Scoville described treatment of cervical spondylosis with bilateral facetectomy and laminectomy as an alternative to anterior cervical decompression described by Cloward and Smith and Robinson (1, 2 and 3). While no instances of C5 palsy were reported in Cloward’s or Smith and Robinson’s original monographs, Stoops and King (4) reported on the findings of neural complications following laminectomy and foraminotomy. Since then, extensive use of both ventral and dorsal procedures to address cervical stenosis, cervical spondylosis, ossification of the posterior longitudinal ligament, and cervical disk herniation has become routine throughout the world. The Japanese literature has predominantly addressed the use of cervical laminectomy and laminoplasty for these pathologies. It is from this body of literature that one can draw most of the knowledge concerning C5 palsy following dorsal cervical intervention (5, 6, 7, 8, 9 and 10). Nonetheless, anterior cervical decompression has also been documented to have an association with the occurrence of C5 palsy (11, 12 and 13).

The incidence of C5 palsy from ventral procedures has been reported to range from 1.6% to 12.1% (11,12). However, the number of reports concerning C5 palsy from ventral procedures is limited to series with comparatively smaller patient numbers. The reported incidence for dorsal procedures ranges from 0% to 30%, with an average incidence of 4.6% (5, 6, 7 and 8,10,13, 14 and 15). Approximately half of patients are reported to demonstrate isolated muscle weakness, while half also report sensory deficits with/without pain in the C5 dermatome distribution. C5 palsy occurs unilaterally in the majority of patients, with approximately 8% of patients demonstrating bilateral findings (7,13). Time of onset is most often within the first week, but delayed presentation has been reported from 2 to 6 weeks postoperatively (5,7,10).

The anatomical course of the C5 ventral nerve rootlets and their relation to the neuroforamina and the inferior vertebral notch may provide insight as to why this is the most commonly affected level. Each spinal nerve root is formed from the rootlets that originate from the dorsal root entry zone and the ventral rootlet exit zone. These rootlets descend from their origination off the spinal cord to a level below the intervertebral foramina and become ensheathed by the dura to form the spinal nerve. This nerve then exits the foramen.

Two studies have examined the angles of the rostral and caudal margins of these rootlets in relation to the spinal cord and their course to the foramen. The rostral rootlet margins of C5 have been reported to range from 23.3 to 45 degrees, and the caudal rootlet margins range from 45.6 to 89 degrees (16,17). By comparison, at the C8 level the superior margin ranges from 19 to 31.6 degrees, while the caudal margin measures 26.5 to 41 degrees. Shinomiya et al. (17) noted that the length of the caudal margin of the ventral rootlet was 6.5 ±1.2 mm at C5 versus 11.6 ± 2.1 mm at C8. Thus, the C5 root has a steeper angle of exit and a shorter length than the more caudal C8 root.

C5 palsy has been attributed to several mechanisms operating together or independently. Sakaura et al. (7) have outlined five possible pathways that may contribute to C5 palsy postoperatively. These include inadvertent injury to the nerve root during surgery, traction of the nerve root from spinal cord shift following decompression, spinal cord ischemia from decreased blood supply, segmental spinal cord dysfunction, or reperfusion injury of the spinal cord.

Interoperative trauma to the nerve root has long been thought to be a primary cause of postoperative C5 palsy. Use of a high-speed burr to drill the bone may cause direct or thermal injury to the nerve root even when great care is exercised near the foramen. Repetitive blunt trauma may also be delivered to the nerve with the use of Kerrison rongeurs during decompression or by use of a nerve hook or other probe to palpate the foramen following decompression. Care should be taken when performing laminoplasty, as the bone of the hinged side may cause occult compression of the nerve root during elevation of the open side. This mechanism, however, does not explain the higher incidence of C5 palsy as compared to other spinal levels.