Study design
This retrospective cohort study was approved by the institutional review board of our hospital (approval number HM21-525), and an opt-out method was used for all potential study participants.
Patient Selection
The included patients underwent SD placement under general anesthesia and skull base reconstruction using a small abdominal fat and pericranial flap. All patients received antibiotics and thromboprophylaxis before surgery. Moreover, as a standard institutional practice, patients were administered 1 g cefazolin intravenously twice daily for 24 h postoperatively. Postoperative rehabilitation was initiated 2 days after surgery.
Study Phases
This study included patients who underwent primary surgery using ATPA between August 1, 2011 and February 28, 2022. No cases were excluded.
This study was conducted in two phases. First, we evaluated the efficacy of postoperative SD placement in CSF leakage prevention. The study period was further divided: first, August 1, 2011 to January 31, 2013 (postoperative SD placement was performed routinely in six patients); second, February 1, 2013 to December 31, 2018 (27 patients, 7 and 20 of whom underwent SD placement and no placement, respectively); and third, January 1, 2019 to February 28, 2022 (all 15 patients, SD was removed immediately after surgery before the waning of anesthesia). To avoid selection bias (as the duration of SD placement was decided by surgeons), we stratified patients into two groups—routine drainage and immediate drain removal groups—to investigate the necessity of postoperative SD placement.
In the second phase, all 48 patients underwent postoperative SD placement and were divided to study whether only 1-day bed rest based on SD placement affects ambulation and muscle disuse: postoperative day (POD) 0 (POD0), POD1, and ≥ POD2 groups, wherein the SD was removed immediately after surgery before the waning of general anesthesia, 1 day postoperatively, and at least 2 days postoperatively, respectively.
CSF leakage was evaluated based on symptoms (postoperative rhinorrhea) and computed tomography (CT) findings (increase or decrease in fluid collection in the antrum, mastoid air cells, and tympanic cavity and presence/absence of pneumocephalus on POD1 and POD7). Subsequently, we evaluated the following parameters in the abovementioned groups.
First, we evaluated intraosseous air cell content (air cells of the petrous apex, direct tracts, and unusual tract on preoperative CT), with or without fluid collection in the mastoid antrum, mastoid air cells, and tympanic cavity on POD1 CT. Direct tract comprises air cells of the petrous apex and/or squamous part of the temporal bone directly connected to the antrum. The unusual tract is composed of air cells of the petrous apex and squamous part of the temporal bone directly connected to the tympanic cavity and attic of the tympanic cavity. Both these tracts were associated with postoperative CSF leakage.
Second, we evaluated the association of medical complications with or without SD placement. Medical complications were defined based on radiographic, culture, prescription, and hospital stay findings. We recorded the presence/absence of medical complications, including pulmonary, gastrointestinal, and urinary complications, as well as deep venous thrombosis.
Patients receiving laxatives were diagnosed with constipation. Presence of deep venous thrombosis on ultrasonography was documented. Patients who required Foley catheter reinsertion and/or medication prescription for urinary disorders were considered to have a urinary complication. Urinary tract infection was defined as a positive urine culture and the administration of antibiotics to the concerned patients. Pulmonary complications were defined as new and worsening oxygen requirements and/or chest radiography showing signs of pneumonia. Finally, we evaluated whether SD placement requiring bed rest prolonged the time to first ambulation and length of hospital stay. Time to first ambulation was defined as the time for a patient to begin walking independently after surgery. Length of hospital stay was defined as the duration until discharge after surgery. All patients were discharged after suture removal.
Statistical analysis
All quantified data except age are expressed as median values (25th percentile value, 75th percentile value). Age is expressed as mean ± standard deviation. Comparisons between the two stratified and three classified groups were performed using the Mann–Whitney U and Kruskal–Wallis tests, respectively. Thereafter, Dunnett’s test was performed to perform multiple comparisons. All statistical analyses were performed using GraphPad Prism (GraphPad Software Inc, San Diego, CA, USA), and P-values < 0.05 were considered to indicate statistical significance.
Skull Base Reconstruction Technique
This skull base reconstruction technique is indicated for evacuating petroclival meningioma, which has attachments from the clivus and petrous apex to the tentorium cerebelli (Fig. 1a). For tumor evacuation, Kawase’s triangle was drilled, the middle cranial fossa dura was cut, and the tentorium cerebelli and posterior fossa dura were removed. Furthermore, air cells of the petrous apex and squamous part of the temporal bone were opened (Fig. 1b). In the first step, an oxidized cellulose sheet for hemostasis was placed on the defected posterior fossa dura to prevent abdominal fat from dropping into the posterior fossa space. Next, the abdominal fat was plugged into the opened air cells of the petrous apex and squamous part of the temporal bone, if possible. This abdominal fat was applied using thrombin and fibrin solutions of exogenous fibrin glue (Fig. 1c). Finally, a pericranial flap was laid on the defected petrous apex and sutured on the middle cranial fossa dura. The space between the subarachnoid space and intraosseous air cells was completely separated to prevent CSF leakage into the nasal cavity via the Eustachian tube. Moreover, the space between the subarachnoid and epidural spaces was separated to prevent CSF leakage into the epidural subcutaneous space.