Correlation between pedicle size and the rate of pedicle screw misplacement in the treatment of thoracic fractures: Can we predict how difficult the task will be?

STUDY DESIGN: A retrospective study. OBJECTIVE: To correlate the incidence of pedicle-screw (PS) misplacement with the dimensions of the pedicles in the treatment of thoracic spine fractures. SUMMARY OF BACKGROUND DATA: The technical challenge of internal fixation with PS in the thoracic spine has been well documented in the literature. However, there are no publications that document the correlation between the pedicle dimensions of the thoracic vertebrae in the preoperative computed tomography scans (CT) and the rate of PS misplacement. METHODS: All patients who had PSs inserted between the T1 and T12 vertebrae during a 24-month period were included in this study. PS position was assessed on high quality CT scans by two independent observers and classified in 2 categories: correct or misplaced. The transverse diameter, craniocaudal diameter and cross-sectional area of the pedicles from T1 to T12 were measured in the pre-operative CT. RESULTS: During the period of this study 36 patients underwent internal fixation with 218 PS. Of the 218 screws, 184 (84.5%) were correct and 34 (15.5%) were misplaced. Misplacement rate was 33% for pedicles with a transverse diameter less than 5 mm, 10.7% for those with a transverse diameter between 5 and 7 mm and 0% for those with a transverse diameter larger than 7 mm. There was a statistically significant difference in the rate of PS misplacement in pedicles with transverse diameter smaller than 5 mm compared with the others. Also, those with transverse diameter between 5.1 and 7 mm compared with those bigger than 7 mm in diameter. The rate of PS misplacement was higher between T3 and T9 (p < 0.05), which in turn correlated with pedicle transverse diameter. CONCLUSION: The rate of PS misplacement in the mid thoracic spine (T4-T9) is high and correlates with pedicle transverse diameter.

The arterial supply of the cervical and thoracic spinal muscles and overlying skin: Anatomical study with implications for surgical wound complications.

Postoperative spinal wound dehiscence is a significant complication following the posterior midline approach. It is postulated that this approach disrupts the vasculature supplying the paraspinal muscles and overlying skin. Although the spinal vasculature has been investigated previously, the smaller arterioles have not been described in the context of the posterior midline approach. Eight cadaveric neck and posterior torso specimens were dissected after injection with a radio-opaque lead oxide mixture and subsequent radiographs taken were analyzed. The deep cervical, vertebral, superficial cervical, and occipital arteries consistently supplied the cervical paraspinal muscles. The latter two arteries also vascularized the overlying skin. The deep cervical arteries were found to be located lateral to the C3 to C6 vertebrae, vulnerable to damage with the posterior approach. In the thoracic region, the superior and posterior intercostal arteries consistently supplied the spinal muscles. In all specimens, two small anastomotic vessels posterior to the laminae were found connecting the intercostal artery perforators. Both the arterial perforators and their anastomotic channels were situated in the surgical field and susceptible to damage with the posterior approach. It is likely that the disruption in spinal vasculature contributes to the multifactorial problem of wound dehiscence with the posterior midline approach.

Single-level/single-stage debridement and posterior instrumented fusion in the treatment of spontaneous pyogenic osteomyelitis/discitis: long-term functional outcome and health-related quality of life.

STUDY DESIGN: Retrospective study. OBJECTIVE: To support single-level posterior debridement and instrumented interbody fusion as a single-stage procedure for spontaneous pyogenic osteomyelitis/discitis. SUMMARY OF BACKGROUND DATA: The best surgical technique for patients with bacterial spinal infections is still a matter of debate. Recent publications suggest that titanium implants can be used safely in the infectious sites in combination with debridement and antibiotic therapy. METHODS: We retrospectively review patients with spontaneous pyogenic osteomyelitis/discitis in whom medical therapy failed, and they consequently underwent posterior decompression and instrumented fusion. Data was collected regarding demographics, clinical presentation, images and laboratory studies, antibiotic treatment, duration of hospitalization, time to achieve radiologic evidence of fusion, postoperative complications, and neurologic function pre- and postoperatively. Quality of life was measured using the EQ5D questionnaire and level of disability with the Oswestry Disability Index. RESULTS: Nine patients, ranging in age from 41 to79 years, with a Frankel score of D in 7 cases and of E in 2 cases, underwent a single-level/single-stage debridement and posterior instrumented fusion with pedicle screws and an interbody and posterolateral autogenous bone graft. Preoperative neurologic deficits improved in all the cases and solid bone fusion was achieved in all 9 patients at 12 months. The mean follow-up period was 67 months. The infection healed after surgery in all the patients and they did not require a second operation to remove the metal implants. Quality of life assessed with the EQ5D questionnaire showed scores ranging between 0.70 and 1. The median Oswestry Disability Index was 15.5%. CONCLUSIONS: These findings support that debridement and posterior instrumented fusion can be performed as a single-stage procedure with no increase in the recurrence rate or morbidity. The outcome has been satisfactory in our patients in terms of the rate of fusion and quality of life.

Factors that predict poor outcomes in patients with traumatic vertebral body fractures.

STUDY DESIGN: Prospective cohort study. OBJECTIVE: To identify factors that predict poor patient-reported outcomes in patients with traumatic vertebral body fracture(s) of the thoracic and/or lumbar spine without neurological deficit. SUMMARY OF BACKGROUND DATA: There is a paucity of information on factors that predict poor patient-reported outcomes in patients with traumatic vertebral body fracture(s) of the thoracic and/or lumbar spine without neurological deficit. METHODS: Patients were identified from the Victorian Orthopaedic Trauma Outcomes Registry (VOTOR). VOTOR includes all patients with orthopaedic trauma admitted to the two adult Level 1 trauma centres in Victoria, Australia. Patient-reported outcomes and data on possible predictive factors, including demographic details, injury-related and treatment-based factors, were obtained from the VOTOR database. Patient-reported outcomes were measured at 12 months post-injury using the 12-Item Short-Form Health Survey (SF-12), a Numerical Rating Scale (NRS) for pain, global outcome questions and data was collected on return to work or study. For the identification of predictive factors, univariate analyses of outcome vs. each predictor were carried out first, followed by logistic multiple regression. RESULTS: 344 patients were eligible for the study and data were obtained for 264 (76.7%) patients at 12 months follow-up. Patients reported ongoing pain at 12 months post-injury (moderate-severe: 33.5%), disability (70.1%) and inability to return to work or study (23.3%). A number of demographic, injury-related and treatment-based factors were identified as being predictive of poor patient-reported outcomes. Patients who had associated radius fracture(s) were more likely to have moderate to severe disability (odds ratio (OR)=3.85, 95% confidence interval=1.30-11.39), a poorer physical health status (OR=3.73, 1.37-10.12) and moderate to severe pain (OR=3.23, 1.22-8.56) at 12 months post-injury than patients without radius fracture. Patients who did not receive compensation for work-related or road traffic-related injuries were less likely to report moderate to severe pain (OR=0.45, 0.23-0.90) or have a poorer mental health status (OR=0.17, 0.04-0.70) at 12 months post-injury than those who received compensation. CONCLUSIONS: The prognostic factors identified in this study may assist clinicians in the identification of patients requiring more intensive follow-up or additional rehabilitation to ultimately improve patient care.

The learning curve of pedicle screw placement: how many screws are enough?

STUDY DESIGN: A retrospective study. OBJECTIVE: Assess the learning curve of pedicle screw (PS) placement of a Spinal Surgery Fellow (SSF) with no previous experience with the technique. SUMMARY OF BACKGROUND DATA: Recent studies have attempted to identify the learning curve for different surgical procedures to define training requirements. Several authors have described a learning curve for PS placement. However, no one has defined the number of PS necessary to be competent in this skill. METHODS: All patients who had PS inserted by the SSF under the supervision of an Attending Spinal Consultant (ASC) and had adequate postoperative radiographs and computed tomography scans available, were included in this study. PS position was assessed by 2 blinded independent observers using a grading scale. PS placement by the SSF was evaluated by examining the assessed position in chronological groups of 40 screws. RESULTS.: Ninety-four patients underwent internal fixation of the spine with 582 PS. Eight cases (40 screws) were excluded because of lack of imaging studies. Of the 542 screws under evaluation, 320 (59%) were performed by the SSF, 187 (34.5%) by the ASC, and 35 (6.5%) by advanced orthopedic or neurosurgical trainees.The rate of misplaced PS performed by the SSF for the first 80 PS was 12.5% and dropped to 3.4% for the remaining 240 screws, which is a statistically significant difference (P < 0.01). Evaluation of computed tomography of vertebrae with PS placed by the SSF on one side and by the ASC on the other showed that the ASC achieved better placement during the first 80 PS (P < 0.01). However, this difference disappeared in the last 240 (P = 1.00). CONCLUSION: The findings demonstrate a learning curve for PS placement. In this series, the asymptote for this technique for an inexperienced SSF, started after about 80 screws (approximately 25 cases).