Predictors of Occurrence and Timing of Post-Tonsillectomy Hemorrhage: A Case-Control Study

Objective: To describe cases and timing of pediatric post-tonsillectomy hemorrhage (PTH), to evaluate predictors of PTH, and to determine the optimal amount of postoperative care unit (PACU) monitoring time. Study Design: Using the Pediatric Health Information System (PHIS) database and electronic medical records, a matched case-control study from 2005 to 2015 was performed. Setting: A single, tertiary-care institution. Subjects and Methods: Each case of PTH was matched with 1 to 4 controls for the following factors: age, sex, surgeon, and time of year. A total of 124 cases of PTH and 479 tonsillectomy controls were included. The rate and timing of postoperative bleeding were assessed, and matched pair analysis was performed using conditional logistic regression. Results: Our institutional PTH rate of 1.9% (130 of 6949) included 124 patients; 15% (19) were primary (≤24 hours), with 50% (9) occurring within 5 hours. Twenty-one percent (4 of 19) of primary PTH patients received operative intervention. Eighty-five percent (105 of 124) of all cases were secondary PTH, and 47% (49) of those patients received operative intervention. Cold steel (OR 1.9, 95% CI 1.1-3.3) and Coblation (OR 1.9, 95% CI 1.2-3.1) techniques and tonsillectomy alone (OR 3.7, 95% CI 1.9-7.2) increased odds of PTH. Patients who developed PTH had 4 times the odds of having a preceding postoperative respiratory event than controls (OR 4.0, 95% CI 1.6-10.0). Conclusion: We conducted a rigorous case-control study for PTH, finding that PTH was associated with use of cold steel and Coblation techniques and with tonsillectomy alone. Patients with a postoperative respiratory event may be more likely to develop a PTH and should be counseled accordingly. A PACU monitoring time of 4 hours is sufficient for outpatient tonsillectomy.