Watertight Sealing without Lumbar Drainage for Incidental Ventral Dural Defect in Transthoracic Spine Surgery: A Retrospective Review of 53 Cases.

STUDY DESIGN:: A retrospective review.

OBJECTIVES:: To evaluate the therapeutic efficacy of the triple layer closure technique to establish watertight sealing without diversion of lumbar drainage, in preventing persistent incidental subarachnoid-pleural fistula and other neurologic complications related to excessive drainage of cerebrospinal fluid (CSF) after dural defect in transthoracic ossified posterior longitudinal ligament (OPLL) surgery.

SUMMARY OF BACKGROUND DATA:: Cerebrospinal fluid (CSF) leakage into the pleural cavity leads to unfavorable conditions for natural healing of incidental durotomy due to the negative pressure environment of the pleural space and lack of wound healing around the bony cavity near the decompressed spinal cord. This often leads to a persistent incideldal subarachnoid-pleural fistula. On the other hand, diversion of lumbar drainage may lead to excessive CSF drainage resulting in intracranial hypotension. To avoid this we studied the efficacy of a modified sealing method to establish a more watertight covering at the ventral dural defect without lumbar CSF drainage.

METHODS:: Fifty-three patients who had CSF leakage from the ventral aspect of the spinal cord during transthoracic spine surgery for thoracic OPLL between 2004 and 2013 were retrospectively reviewed. Patients were divided into two groups: a conventional group (Group A) and a triple layer closure group (Group B). In Group A(n=33patients), the dural defect was covered with fibrin glue (Beriplast P, CSL Behring, Germany) mixed with gelfoam (Spongostan Standard, Ethicon, Somerville, USA) with subsequent subarachnoid lumbar drainage. In Group B(n=20 patients), the dural defect was sealed using the triple layer technique with two layers of fibrin glue and gelatin sponge plus a third layer of synthetic hydrogel (Duraseal, Dural Sealant System, Covidien Ilc., USA) without subsequent subarachnoid lumbar drainage. Both groups had chest tubes, which drained through an underwater seal. Clinical data included duration and total amount of drainage (chest tube and lumbar drainage), related complications, and duration of hospital stay were compared between the two groups.

RESULTS:: Compared to the patients in Group A, Group B had a significantly smaller total volume of drainage and shorter chest tube drainage time (P<0.05) during their hospital stay. In Group A, complications occurred in 6 cases (18.2%) including 3 cases of intracranial hypotension combined with transient mental status alteration, postural headache, and dizziness, one case of regional atelectasis with pneumonia and 2 cases of revision thoracotomy. Revision thoracotomy was performed to treat persistent subarachnoid-pleural fistula due to significant and prolonged CSF leakage. In group B, there were no complications and no revision thoracotomy was needed. The mean duration of hospital stay was shorter in group B(15.6 d) than group A (22.4 d).

CONCLUSION:: The established watertight closure of the dural defect using the triple layer sealing method without lumbar drainage was more effective and safe.