Epidemiology of late postoperative bleeding in OSA-related tonsil surgery: a multicentric retrospective study.

Abstract

Introduction: Tonsil-related procedures are considered fundamental and effective in the surgical treatment of obstructive sleep apnea (OSA). The range of techniques includes intratonsillar approaches, such as tonsillotomy (TT), as well as extracapsular procedures, such as tonsillectomy (TE) and uvulopalatopharyngoplasty (UPPP). Patients undergoing these procedures span all age groups, from children to seniors.

Methods: This multicentric retrospective study, conducted between 2014 and 2018, analysed data from 3,498 patients who underwent bilateral TT, TE, or UPPP for OSA or ronchopathy. The cohort included 2,221 men (63.49%) and 1,277 women (36.51%). Of these, 2,808 patients (80.27%) underwent TT, 226 (6.46%) underwent TE, and 464 (13.26%) underwent UPPP.

Results: Late postoperative haemorrhage (LPOH) occurrence was significantly associated with the type of surgery (p < 0.0001) and the hospital where the procedure was performed (p < 0.0001). The incidence of LPOH in the TT group ranged from 0% to 5.88% across hospitals (p = 0.0068); whereas in the TE and UPPP groups, rates ranged from 0% to 33.33% (p = 0.0413 and p = 0.0409, respectively). The occurrence of repetitive bleeding was not influenced by treatment choice (readmission vs. outpatient care, observation vs. surgical revision, general vs. local anaesthesia). The severity of bleeding in all three groups was not affected by age and gender. The use of anticoagulants negatively impacted LPOH severity (p = 0.0166) in the UPPP group. No deaths occurred in our sample; however, three cases of severe postoperative bleeding (grade "D") were observed.

Conclusion: Late postoperative haemorrhage remains a serious complication of tonsil-related surgery with the potential for life-threatening outcomes. The marked variability in bleeding incidence between surgical techniques and departments highlights the need for standardised perioperative protocols. Although no fatalities occurred, the occurrence of severe cases underlines the importance of vigilant postoperative monitoring. In our OSA cohort, tonsillotomy showed favourable safety, and recent evidence suggests it may represent a valuable alternative also in recurrent tonsillitis, warranting further research.