Objectives-To characterise the pattern of and risk factors for degenerative changes of the cervical spine in patients with spasmodic torticollis and to assess whether these changes aVect outcome after selective peripheral denervation. Methods-Preoperative CT of the upper cervical spine of 34 patients with spasmodic torticollis referred for surgery were reviewed by two radiologists blinded to the clinical findings. Degenerative changes were assessed for each joint separately and rated as absent, minimal, moderate, or severe. Patients were clinically assessed before surgery and 3 months postoperatively by an independent examiner using standardised clinical rating scales. For comparison of means a t test was carried out. To determine whether an association exists between the side of degenerative changes and type of spasmodic torticollis a 2 test was used. Changes in severity, disability, and pain before and after surgery were calculated using a Wilcoxon matched pairs signed ranks test. Results-Fourteen out of 34 patients had moderate or severe degenerative changes. They were predominantly found at the C2/C3 and C3/C4 level and were significantly more likely to occur on the side of the main direction of the spasmodic torticollis (p=0.015). There was no significant diVerence in age, sex, duration of torticollis, overall severity, degree of disability, or pain between the group with either no or minimal changes and the group with moderate or severe changes. However, in the second group the duration of inadequate treatment was longer (10.1 v 4.8 years; p=0.009), head mobility was more restricted (p=0.015), and head tremor was more severe (p=0.01). At 3 months postoperatively, patients with no or minimal degenerative changes showed a significant improvement in pain and severity whereas no diVerence was found in those with moderate or severe changes. Conclusions-Patients with spasmodic torticollis have an increased risk of developing premature degenerative changes of the upper cervical spine that tend to be on the side towards which the head is turned or tilted and compromise outcome after surgery. EVective early treatment of spasmodic torticollis with botulinum toxin seems to have a protective eVect. Patients with spasmodic torticollis and restricted head mobility who do not adequately respond to treatment should undergo imaging of the upper cervical spine. Patients with imaging evidence of moderate or severe degenerative changes seem to respond poorly to selective peripheral denervation. (J Neurol Neurosurg Psychiatry 2000;68:465-471)

Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.

Level III, comparative series.

Background: Several operative interventions are available to alleviate pain in hallux rigidus, and the optimal operative technique is still a topic of debate among surgeons. Three of these are arthrodesis, cheilectomy, and Keller’s arthroplasty. Currently, it is unclear which intervention yields the best long-term result. The aim of this study was to assess which of these interventions performed best in terms of patient-reported outcome, pain scores, and disease recurrence at long-term follow-up. Methods: These data are the follow-up to the initial study published in 2006. In the original study, 73 patients (n = 89 toes) with symptomatic hallux rigidus were recruited and underwent first metatarsophalangeal joint arthrodesis (n = 33 toes), cheilectomy (n = 28 toes), or Keller’s arthroplasty (n = 28 toes). Outcome measures were AOFAS hallux metatarsophalangeal-interphalangeal (HMI) score, and pain was assessed with a visual analog scale (VAS) at a mean follow-up period of 7 years. Patients of the original study were identified and invited to participate in the current study. Data were collected in the form of AOFAS-HMI score, VAS pain score, Manchester-Oxford Foot Questionnaire (MOXFQ), and Forgotten Joint Score (FJS-12). In addition, a clinical examination was performed and radiographs were obtained. Data were available for 37 patients (45 toes), with a mean follow-up period over 22 years. Results: AOFAS-HMI and VAS pain score improved during follow-up only in arthrodesis patients. Furthermore, no statistically significant differences in clinical and patient-reported outcome were detected between groups based on AOFAS-HMI, VAS pain, MOXFQ, or FJS-12. However, clinically important differences in patient-reported outcomes and pain scores were detected, favoring arthrodesis. Radiographic disease progression was more evident after cheilectomy compared with Keller’s arthroplasty. Conclusion: Arthrodesis, cheilectomy, and Keller’s arthroplasty are 3 sucessful operative interventions to treat symptomatic hallux rigidus. Because clinically important differences were detected and symptoms still diminish many years after surgery, a slight preference was evident for arthrodesis. Level of Evidence: Level III, comparative study.