Role of Peri- and Postoperative Antibiotics in Microscopic Laryngeal Surgery

There are no widely accepted guidelines for the use of peri- and postoperative antibiotics for microdirect laryngoscopy (MDL) to prevent postoperative surgical site infection (SSI). This study examines the incidence of not only SSI, but also respiratory and oropharyngeal infections among patients who underwent MDL.

Retrospective cohort.

Medical records from a quaternary laryngology practice were reviewed retrospectively for patients who underwent MDL between October 2015 and December 2018. Adults who underwent MDL with excision of nonmalignant masses, vocal fold (VF) reconstruction with buccal graft, superficial VF injection, VF injection augmentation, use of carbon dioxide (CO₂) or potassium titanyl phosphate (KTP) laser, and laryngeal dilation were included. Subjects were divided into three groups based on the use of antibiotics during the peri- or postoperative period. Group 1 did not receive peri- or postoperative antibiotics. Group 2 received perioperative antibiotics without postoperative antibiotics. Group 3 received both peri- and postoperative antibiotics. Antibiotic use was determined by surgeon policy, which changed over time, not by patient risk factors. Other data collected include demographics, medical comorbidities, surgical procedure, operative time, American Society of Anesthesiology class, incidence of SSI, respiratory or oropharyngeal infection, and postoperative antibiotic and steroid prescriptions for complications.

190 surgeries met inclusion criteria. Groups 1, 2, and 3 consisted of 113, 57, and 20 subjects, respectively. No SSI were reported. Baseline characteristics of median age, American Society of Anesthesiology, wound class, and gender did not differ between groups. Groups were found to have differed in type of procedure performed, with a predominance of vocal fold excision and superficial injection in Group 1, injection laryngoplasty using autologous fat graft and laryngeal reconstruction in Group 3, and a combination of procedure types in Group 2. Laser treatment with CO₂ and/or KTP was utilized in 35% of procedures. The mean operative times (total time in which the laryngoscope remained in the oral cavity) for Groups 1, 2, and 3 were 29 (SD = 13), 42 (SD = 23), and 73 minutes (SD = 78), respectively; this was significantly different among the three groups (P < 0.0001). By logistic regression models, operative time was not significantly associated with the overall use of antibiotics (Χ² = 1.81, P = 0.18) or corticosteroid (Χ² = .05, P = .82) prescriptions during follow-up. In Groups 1, 2, and 3, 11%, 14%, and 10% of patients, respectively, required a follow-up antibiotic prescription for non-SSI and 11%, 12%, and 0% of patients, respectively, required a follow-up corticosteroid prescription within 30 days postoperatively. There was no significant difference in patients with diabetes or pulmonary comorbidity and patients without comorbidity in use of postoperative antibiotic (P = .45) or corticosteroid (P > .99) prescription. Within 30 days of follow-up, patients in Group 1 experienced VF hemorrhage (defined as any subepithelial blood in the area of excision) at an incidence of 19%. Complications within 30 days of follow-up showed an incidence of substantial pain ranging between 2%-4% and respiratory or oropharyngeal infection between 10%-12%. Two patients in Group 2 reported emergency department visits for dyspnea. One patient in Group 3 was admitted for tongue and floor of mouth swelling secondary to prolonged suspension.

There were no incidents of SSI in this cohort. The need for follow-up antibiotic and/or corticosteroid prescription did not differ in relation to use or non-use of perioperative antibiotics, and neither did the incidence of respiratory or oropharyngeal infection. Longer operative time was associated with both peri- and postoperative antibiotic use, but there is no evidence yet that antibiotic use was necessary. These findings suggest that avoidance of routine perioperative and/or postoperative antibiotics may be appropriate; and further research should be encouraged.