Object: The authors report a series of eight consecutive cases in which epidural abscesses in the cervical spine were treated by microsurgery without arthrodesis, including two cases of concomitant pyogenic and tubercular infection. Methods: The authors used a minimally invasive surgical approach consisting of single-level anterior microsurgical discectomy and drainage of the epidural abscess via a silicone catheter, and then initiated antibiotic therapy. At follow-up examination (mean duration 39 months), six patients exhibited complete recovery and two suffered from minor residual deficits. In all cases, spontaneous vertebral fusion occurred. Sagittal alignment was maintained in seven patients, and in one there was slight asymptomatic kyphosis. In two patients, tubercular and pyogenic infections were found. Prior intervention for dental infection was recorded in four cases. Conclusions: In the absence of preoperative spinal instability, microsurgical drainage of the abscess followed by specific antibiotic therapy resulted in spinal cord decompression and neurological recovery, thereby facilitating spontaneous fusion and vertebral stability. The presence of combined tubercular and pyogenic infections of the cervical spine should be considered, especially in patients whose immune systems are depressed.

Needle EMG may be negative in mild or predominantly sensory lumbosacral radiculopathies. In such cases, an increase in the latency of the soleus H-reflex is a useful diagnostic criterion for establishing sensory fiber compromise at the S1 root level. However, if clinical signs of radicular involvement are lacking, the latency of the H-reflex is normal. We therefore studied the recruitment curve of the soleus H-reflex to investigate whether a change in the electrical threshold for eliciting the H-reflex might be a more sensitive criterion for detecting subclinical S1 root dysfunction. Clinical and electrophysiological findings from 26 patients with chronic back pain and radiculopathy were compared with data obtained from 40 healthy subjects. An increase in the mean H-reflex threshold was the only abnormal electrophysiological finding in patients with no clinical sign of root injury (58%). A decrease in the mean H-reflex amplitude and a prolongation of H-reflex latency was observed in patients with radicular signs (42%). In both patients groups, F-wave and needle EMG studies were normal. No radiological evidence of S1 root compression was found. The study of the recruitment curve of the soleus H-reflex may be usefully associated to F-wave and needle EMG studies to detect possible S1 root dysfunction in mild lumbosacral radiculopathies. An increase in H-threshold may be the earliest abnormality in the absence of focal neurological signs.