Objectives: To analyze the incidence and severity of postoperative hemorrhage following transoral robotic surgery (TORS) for oropharyngeal carcinoma since initiating the practice of preoperative prophylactic transcervical arterial ligation.

Design: Cross-sectional retrospective outcomes review from October 19th, 2012 to October 2nd, 2015.

Setting: Academic medical center

Methods: The medical records of 177 consecutive patients who underwent transoral robotic surgery for carcinoma of the tonsil or tongue base performed by a single head and neck surgeon were reviewed. Patients with a history of prior head and neck radiation therapy and those who underwent simultaneous free tissue reconstruction were excluded.  Patients treated with TORS for benign disease or laryngeal or parapharyngeal space lesions were also excluded. Each subject underwent neck dissection with ligation of the ipsilateral lingual, facial and superior thyroid arteries prior to TORS (mean 7.8±0.5 days). For each subject, age, gender, dates of surgery, tumor location and TNM classification were extracted. For patients who experienced postoperative bleeding, details including associated morbidity, readmission and return to operating room were also extracted. The Mayo Clinic classification system for postoperative hemorrhage was used to classify each bleeding episode as minor, intermediate, major or severe.

Results: A total of 145 patients were eligible for the study.  Fourteen patients (9.7%) experienced postoperative bleeding at an average of 7.6 ± 1.3 days after TORS. Of the 14 patients with oropharyngeal hemorrhage, 7 were classified as minor, 6 intermediate and 1 major. Seven patients (50%) who experienced post-operative bleeding required operative intervention with direct laryngoscopy and cautery. No patients experienced severe or life-threatening bleeds, defined as bleeding resulting in hypoxia/airway compromise requiring tracheostomy, cardiopulmonary arrest, or hemodynamic instability requiring blood transfusion. Postoperative hemorrhage was significantly associated with male gender (p=0.049) and younger age (p=0.049). Patients with T3 or T4a primary tumors were at increased risk of postoperative hemorrhage compared to those with T1 or T2 primary tumors (RR = 4.1, 95% CI: 1.4, 12.2).  There was no association between postoperative hemorrhage and primary tumor location (p=0.36) or time interval between neck dissection and robotic surgery (p=0.81).

Conclusions: Prior studies have established the safety and efficacy of TORS in the treatment of early-stage malignant tumors of the oropharynx. However, postoperative hemorrhage remains a significant concern with previously published hemorrhage rates ranging between 1.5%-11.5%. We present the largest series to date of patients undergoing TORS for oropharyngeal carcinoma after prophylactic transcervical arterial ligation.  While the incidence of postoperative oropharyngeal bleeding in this series is comparable to previously reported rates, there have been no major or severe bleeds since prophylactic arterial ligation became routine at our institution. This is in contrast to a recently published study of 224 TORS patients in which 4.5% of patients had major and 2.2% of patients experienced severe postoperative hemorrhage. Our findings suggest that prophylactic arterial ligation may reduce the risk of major and severe postoperative hemorrhage and increase the safety of TORS.