Object. The aims of this study were to review the incidence of cerebrospina
l fluid (CSF) leakage complicating the removal of acoustic neuroma. and to
identify factors that influence its occurrence and treatment.
Methods. Prospective information on consecutive patients who underwent oper
ation for acoustic neuroma was supplemented by a retrospective review of th
e medical records in which patients with CSF leaks complicating tumor remov
al were identified. This paper represents a continuation of a previously pu
blished series and thus compiles the authors' continuous experience over th
e last 24 years of practice.
In 624 cases of acoustic neuroma. the authors observed an overall incidence
of 10.7% for CSF leak. The rate of leakage was significantly lower in the
last 9 years compared with the first 15, most likely because of the abandon
ment of the combined translabyrinthine (TL)-middle fossa exposure. There wa
s no difference in the leakage rate between TL and retrosigmoid (RS) approa
ches, although there were differences in the site of the leak (wound leaks
occurred more frequently after a TL and otorrhea after an RS approach, resp
ectively). Tumor size (maximum extracanalicular diameter) had a significant
effect on the leakage rate overall and for RS but not for TL procedures. T
he majority of leaks ceased with nonsurgical treatments (18% with expectant
management and 49% with lumbar CSF drainage). However, TL leaks (especiall
y rhinorrhea) required surgical repair significantly more often than RS lea
ks. This has not been reported previously.
Conclusions. The rate of CSF leakage after TL and RS procedures has remaine
d stable. Factors influencing its occurrence include tumor size but not sur
gical approach. The TL-related leaks had a significantly higher surgical re
pair rate than RS-related leaks, an additional factor to consider when choo
sing an approach. The problem of CSF leakage becomes increasingly important
as nonsurgical treatments for acoustic neuroma are developed.