Predictive clinical exam findings in post-tonsillectomy hemorrhage

Post tonsillectomy hemorrhage (PTH) is a common complication of tonsillectomy. Our objectives were to: 1) Examine the postoperative course of patients presenting to St. Christopher's Hospital for Children (SCHC) with PTH; 2) Compare patients with and without a blood clot visualized in the tonsillar fossa at time of presentation to determine if outcomes regarding return to the operating room (OR) differ.

This was a retrospective chart review conducted at an academic, tertiary, pediatric hospital in an urban setting. Pediatric patients who underwent a tonsillectomy with concurrent adenoidectomy and were admitted for observation following secondary post tonsillectomy hemorrhage were reviewed. The effects of age, gender, indication, and clinical exam findings on admission on the rate of eventual return to the OR for control of hemorrhage were also analyzed. Chi-square analysis and Fisher's exact test were used to compare the significance of categorical frequencies.

The rate of blood clot presence in our cohort was 50.9% (28/55). Return to OR rates were defined as patients who began actively hemorrhaging following admission for observation, further stratified by presence or absence of clot on admission physical exam. There was a statistically significant higher rate of return to OR in patients who presented with a clot (46.6%) on clinical exam versus no clot (18.5%) after resolved post tonsillectomy hemorrhage (p < .027). Furthermore, patients with a blood clot present were significantly more likely to require OR sooner (21.31 h from admission) than those without a clot (100.75 h from admission) (p < .012). There was no statistically significant higher rate of blood clot presence or rate of return to OR in groups based on age, gender, or indication.

Pediatric patients presenting after resolved secondary PTH with a blood clot visualized in the tonsillar fossa are more likely to require return to the OR for hemostasis and cautery than are those without a blood clot, and this is more likely to occur within 24 h of admission. Thus, patients with a blood clot on initial presentation may benefit from admission for a 24-h observation period, while a similar observation period may be unproductive for patients without a blood clot.

Patients who present with a resolved secondary PTH and a blood clot present on clinical exam require return to the OR more often than patients presenting without a blood clot. While previously controversial, we feel that this demonstrates that a 24-h observation of a patient with a clot on exam is reasonable.