No abstract

OBJECTIVE The vascularized nasoseptal flap (NSF) has become the workhorse for skull base reconstruction during endoscopic endonasal surgery (EES) of the ventral skull base. Although infrequently reported, as with any vascularized flap the NSF may undergo ischemic necrosis and become a nidus for infection. The University of Pittsburgh Medical Center's experience with NSF was reviewed to determine the incidence of necrotic NSF in patients following EES and describe the clinical presentation, imaging characteristics, and risk factors associated with this complication. METHODS The electronic medical records of 1285 consecutive patients who underwent EES at the University of Pittsburgh Medical Center between January 2010 and December 2014 were retrospectively reviewed. From this first group, a list of all patients in whom NSF was used for reconstruction was generated and further refined to determine if the patient returned to the operating room and the cause of this reexploration. Patients were included in the final analysis if they underwent endoscopic reexploration for suspected CSF leak or meningitis. Those patients who returned to the operating room for staged surgery or hematoma were excluded. Two neurosurgeons and a neuroradiologist, who were blinded to each other's results, assessed the MRI characteristics of the included patients. RESULTS In total, 601 patients underwent NSF reconstruction during the study period, and 49 patients met the criteria for inclusion in the final analysis. On endoscopic exploration, 8 patients had a necrotic, nonviable NSF, while 41 patients had a viable NSF with a CSF leak. The group of patients with a necrotic, nonviable NSF was then compared with the group with viable NSF. All 8 patients with a necrotic NSF had clinical and laboratory evidence indicative of meningitis compared with 9 of 41 patients with a viable NSF (p < 0.001). Four patients with necrotic flaps developed epidural empyema compared with 2 of 41 patients in the viable NSF group (p = 0.02). The lack of NSF enhancement on MR (p < 0.001), prior surgery (p = 0.043), and the use of a fat graft (p = 0.004) were associated with necrotic NSF. CONCLUSIONS The signs of meningitis after EES in the absence of a clear CSF leak with the lack of NSF enhancement on MRI should raise the suspicion of necrotic NSF. These patients should undergo prompt exploration and debridement of nonviable tissue with revision of skull base reconstruction.

OBJECTIVE Surgical site infections (SSIs) are a major source of morbidity after spinal surgery. Several recent studies have described the finding that applying vancomycin powder to the surgical bed may reduce the incidence of SSI. However, applying vancomycin in high concentrations has been shown in vitro to inhibit osteoblast proliferation and to induce cell death. Vancomycin may have a deleterious effect on dural healing after repair of an intentional or unintentional durotomy. This study was therefore undertaken to assess the effect of different concentrations of vancomycin on a human dura mater cell culture. METHODS Human dura intended for disposal after decompressive craniectomy was harvested. Explant primary cultures and subcultures were subsequently performed. Cells were characterized through common staining and immunohistochemistry. A growth curve was performed to assess the effect of different concentrations of vancomycin (40, 400, and 4000 μg/ml) on cell count. The effect of vancomycin on cellular shape, intercellular arrangement, and viability was also evaluated. RESULTS All dural tissue samples successfully developed into fusiform cells, demonstrating pseudopod projections and spindle formation. The cells demonstrated vimentin positivity and also had typical features of fibroblasts. When applied to the cultures, the highest dose of vancomycin induced generalized cell death within 24 hours. The mean (± SD) cell counts for control, 40, 400, and 4000 μg/ml were 38.72 ± 15.93, 36.28 ± 22.87, 19.48 ± 6.53, and 4.07 ± 9.66, respectively (p < 0.0001, ANOVA). Compared with controls, vancomycin-exposed cells histologically demonstrated a smaller cytoplasm and decreased pseudopodia formation resulting in the inhibition of normal spindle intercellular arrangement. CONCLUSIONS When vancomycin powder is applied locally, dural cells are exposed to a concentration several times greater than when delivered systemically. In this in vitro model, vancomycin induced dural cell death, inhibited growth, and altered cellular morphology in a concentration-dependent fashion. Defining a safe vancomycin concentration that is both bactericidal and also does not inhibit normal dural healing is necessary.