To determine whether the choice of surgical approach affects the rate of postoperative cerebrospinal fluid leakage in patients who have undergone surgical resection of acoustic neuroma.
Retrospective chart review.
Tertiary referral center.
Three hundred patients who underwent surgery for acoustic neuromas were selected by consecutive medical record number until 100 resections via each surgical approach (translabyrinthine, middle fossa, and retrosigmoid) had been gathered.
Main Outcome Measures
Surgical approach used, cerebrospinal fluid leak incidence, tumor size, patient age.
Postoperative cerebrospinal fluid leak of any severity was observed in 13% of translabyrinthine, 10% of middle fossa, and 10% of retrosigmoid patients. These difference in the rate of cerebrospinal fluid leakage were not statistically significant (p = 0.82). The majority of leaks were managed conservatively with fluid and activity restriction, often accompanied by a period of lumbar subarachnoid drainage. There was a need to return to the operating room for a definitive procedure in 4% of translabyrinthine, 2% of middle fossa, and 3% retrosigmoid patients; again not statistically different among the approaches (p = 0.43). Tumor size was not correlated with cerebrospinal fluid leak rate (p = 0.13). Patient age, for patients older than 50 years, was suggestive of increased odds of cerebrospinal fluid leak (p = 0.06).
Neither surgical approach nor tumor size affects the rate of postoperative cerebrospinal fluid leakage or the necessity of managing a leak with a return to the operating room. Cerebrospinal fluid leakage rates have remained stable in recent decades despite numerous innovative attempts to improve dural closure, seal transected air cell tracts, and occlude anatomic pathways. The finding that leak rates were similar among three dissimilar surgical techniques suggests that factors other than techniques of wound closure, such as transient postoperative rises in cerebrospinal fluid pressure, may be responsible for these recalcitrant cases.