Background: There are no current guidelines on the timing of resuming nocturnal non-invasive positive pressure ventilatory therapy after skull base surgery. In patients with obstructive sleep apnea (OSA) undergoing either open or endoscopic skull base surgery, an analysis of risk factors and timing of postoperative cerebrospinal fluid (CSF) leaks would provide important prognostic information and may help clinicians determine when to restart continuous positive airway pressure (CPAP) therapy. There are case reports of CSF leaks and symptomatic pneumocephalus after resumption of CPAP postoperatively, but these appear to be rare although serious events1,2.
Methods: A retrospective review of patients at a single institution between 2002 and 2015 with a diagnosis of OSA and age/gender matched patients without a diagnosis of OSA undergoing skull base surgery who experienced a postoperative CSF leak was performed.
Results: Ten patients were identified with OSA and a postoperative CSF leak. Those with OSA developed a postoperative CSF leak following skull base surgery at an average of 4.7 days (range: 1-19 days) compared to ten patients without OSA who developed a postoperative CSF leak at an average of 6.9 days (range: 1-20 days) (p=0.42). Seven of the ten patients with OSA developed CSF leaks within 72 hours of the surgery compared to four of ten patients without OSA. No patients with OSA used CPAP in the postoperative period. Nine of ten CSF leaks in the OSA group developed after primary skull base surgery, while one occurred after primary attempted repair of a leak. Four of ten CSF leaks in the non-OSA group developed after primary skull base surgery, with the remainder occurring after primary attempted repair of a leak. Average BMI for the OSA group was 39.3 kg/m2 compared to 28.5 kg/m2 for the group without OSA (p<0.01). Average age for both groups was 52.8 years. Location of the leak was most commonly located in the sellar region or sphenoid sinus in both groups. The most common type of repair in the OSA group was an autologous fat graft, while in the non-OSA group it was an autologous fat graft combined with a nasoseptal flap.
Conclusion: There is a non-significant trend towards early onset of CSF leak in patients with OSA compared to age/gender matched patients without OSA. As CPAP was not utilized during this period, this difference cannot be attributed to its use. A larger case-control study would be beneficial to determine if there is a definitive relationship between OSA and postoperative skull base CSF leaks, perhaps due to higher airways pressures or other sequelae of the most-common underlying risk factor of OSA, obesity. The use of nasoseptal flaps in these higher risk populations would be useful to prevent CSF leaks, and could permit earlier resumption of CPAP given its benefits.
1 White-Dzuro, Gabrielle A., et al. "Risk of post-operative pneumocephalus in patients with obstructive sleep apnea undergoing transsphenoidal surgery." Journal of Clinical Neuroscience 29 (2016): 25-28.
2 Kopelovich, Jonathan C., et al. "Pneumocephalus with BiPAP use after transsphenoidal surgery." Journal of clinical anesthesia 24.5 (2012): 415-418.