Discontinuation of Postoperative Prophylactic Antibiotics for Endoscopic Endonasal Surgery

Direct access through the sinuses and nasopharyngeal mucosa in the endoscopic endonasal approach (EEA) raises concern for a contaminated operative environment and subsequent infection. The reported rate of meningitis in endoscopic endonasal skull base surgery in the literature ranges from 0.7 to 3.0% [1, 2]. The only factor identified as being independently associated with meningitis in a statistically significant manner is cerebrospinal fluid (CSF) leak [1–5]. However, many centers performing high volume of EEAs use postoperative antibiotic coverage independent of the presence intraoperative or postoperative CSF leak. Furthermore, while meningitis remains a severe concern, most centers use postoperative gram-positive coverage to prevent toxic shock syndrome caused by Staphylococcus aureus infection in the setting of prolonged nasal packing. There are currently a multitude of approaches regarding perioperative antibiotic coverage in EEAs [1–4]. Given the lack of consensus in the literature and our experience regarding the benefit of discontinuation of prolonged prophylactic antibiotics throughout the breadth of neurosurgical procedures, we sought to analyze the need for postoperative antibiotics in EEAs further. As such, we performed a prospective analysis compared with a retrospective cohort to delineate whether discontinuation of postoperative antibiotics leads to a change in the rate of postoperative infections.

The retrospective cohort consisted of patients who underwent an EEA from January 1, 2013 to May 31, 2019. These patients all received postoperative antibiotics while nasal packing was in place (median 7 days). Starting on April 1, 2019 until August 1, 2019, we discontinued postoperative antibiotic use. Patients from this group made up the prospective cohort. The retrospective cohort had 315 patients (66% pituitary macroadenomas vs. 7% microadenomas, 4% meningiomas, 4% craniopharyngiomas, 4% chordomas, and 15% others) while the prospective group had 23 patients (57% pituitary macroadenomas, 30% craniopharyngiomas, 8% meningiomas/chordomas, and 5% others). The primary endpoint was rate of postoperative infections and specifically, meningitis and multidrug resistant organism (MDRO) infections.

There was no statistically significant difference in the use of nasal packing (p = 0.085), intraoperative CSF leak (p = 0.133), and postoperative CSF leak (p = 0.507) between the two groups. There was also no significant difference in the number of patients with positive preoperative MSSA and MRSA nasal swabs (p = 0.622). There was a significant decrease in the number of patients discharged with antibiotics (55.1% in the retrospective and 4.5% in the prospective group, p = 0.000). The number of patients with positive blood cultures (p = 0.701) and positive urine cultures (p = 0.691) did not differ significantly between the two groups. Finally, there was no statistically significant difference in postoperative CSF infections (p = 0.34) or MDRO infections (0.786) between the two groups.

We describe promising preliminary results that demonstrate that discontinuation of postoperative antibiotics in EEAs do not lead to a statistically significant increase in the rate of postoperative CSF or MDRO infections. The previous algorithm for postoperative antibiotic coverage in our center, like many centers, called for gram-positive coverage, which may have contributed to the overall preponderance of gram-negative meningitis cases in this cohort.