Search Results

You are looking at 1 - 3 of 3 items for

  • Author or Editor: Steven Knafo x
  • All content x
Clear All Modify Search
Restricted access

Steven Knafo, Charles Court, and Fabrice Parker


Spinal deformity after surgery for intramedullary tumors is a potentially serious complication that may require subsequent fusion. The aim of this study was to determine whether there were risk factors that could be used to predict postoperative sagittal deformity.


The authors conducted a retrospective study of patients harboring an intramedullary tumor who had undergone surgery at a single center between 1985 and 2011. The main outcome of interest was the difference, at the last follow-up, between post- and preoperative measures of the Cobb angle formed by the superior and inferior limits of the laminectomy (ΔCobb).


Sixty-three patients were eligible for inclusion in the study. The mean sagittal deformity, measured as described above, was 15.9° (range 0°–77°) at a mean follow-up of 85.4 months (range 4–240 months). Univariate analysis showed increased sagittal deformity in patients 30 years old or younger (21.9° vs 13.7°, p = 0.04), undergoing a laminectomy involving 4 or more levels (19.3° vs 12.1°, p = 0.04), and undergoing a laminectomy that included a spinal junction (20.8° vs 12.4°, p = 0.02). Multivariate analysis showed that only age (p = 0.01) and the number of spinal levels involved in the laminectomy (p = 0.014) were significant and independent predictors of postoperative sagittal deformity. The linear regression equation drawn from this model allows one to quantitatively predict sagittal deformity for any follow-up time point after surgery.


Authors of this study developed a statistical tool that could be used to plan surgery and follow-up as regards the risk of sagittal spinal deformity in patients undergoing surgery for intramedullary tumors.

Restricted access

Steven Knafo, Fabrice Parker, Anne Herbrecht, Charles Court, and Guillaume Saliou

Subarachnoid-pleural fistula is a well-described complication after anterior surgery for thoracic disc herniation, but is difficult to treat by means of traditional chest and lumbar drains due to interference by positive ventilation pressures that may keep the fistula open and prevent proper closure. Current treatment strategies include surgical repair, which is technically challenging, and noninvasive positive pressure ventilation, which can take several weeks to be effective. In this report, the authors describe a novel treatment for subarachnoid-pleural fistula using percutaneous obliteration with Onyx.

Surgery for removal of a T7–8 disc herniation associated with ossification of the posterior longitudinal ligament was performed in a 56-year-old woman via an anterior transthoracic transpleural approach. Ten days after surgery, she presented with diplopia due to a subarachnoid-pleural fistula that was confirmed by CT myelography. Percutaneous injection of Onyx was performed under local anesthesia. Postprocedure CT showed complete obliteration of the fistula with no adverse events. A CT scan obtained 1 month later showed complete resolution of the pleural effusion. Neurological examination at 3 months postsurgery was normal. Clinical and radiological follow-up at 1 year showed complete recovery and no sign of fistula recurrence. Percutaneous treatment for subarachnoid-pleural fistula is an easy, safe, and effective strategy and can therefore be proposed as a first-line option for this challenging complication.

Restricted access

Gonzague Guillaumet, Nozar Aghakhani, Silvia Morar, Razvan Copaciu, Fabrice Parker, and Steven Knafo


Surgical treatment for nonforaminal syringomyelia related to spinal arachnoiditis is still controversial. The authors sought to assess respective outcomes and rates of reintervention for shunting and spinal cord untethering (arachnolysis) in spinal arachnoiditis with syringomyelia.


This retrospective cohort study was conducted at a single reference center for syringomyelia. Patients undergoing arachnolysis and/or shunting interventions for nonforaminal syringomyelia were screened.


The study included 75 patients undergoing 130 interventions. Arachnolysis without shunting was performed in 48 patients, while 27 patients underwent shunting. The mean follow-up between the first surgery and the last outpatient visit was 65.0 months (range 12–379 months, median 53 months). At the last follow-up, the modified McCormick score was improved or stabilized in 83.4% of patients after arachnolysis versus 66.7% after shunting. Thirty-one (41.3%) patients underwent reintervention during follow-up, with a mean delay of 33.2 months. The rate of reintervention was 29.2% in the arachnolysis group versus 63.0% in the shunting group (chi-square = 8.1, p = 0.007). However, this difference was largely driven by the extension of the arachnoiditis: in patients with focal arachnoiditis (≤ 2 spinal segments), the reintervention rate was 21.6% for arachnolysis versus 57.1% for shunting; in patients with extensive arachnoiditis, it was 54.5% versus 65.0%, respectively. Survival analysis assessing the time to the first reintervention demonstrated a better outcome in both the arachnolysis (p = 0.03) and the focal arachnoiditis (p = 0.04) groups.


Arachnolysis led to fewer reinterventions than shunting in patients with nonforaminal syringomyelia. There was a high risk of reintervention for patients with extensive arachnopathies, irrespective of the surgical technique.