Is the Risk of Clinically Significant Hemorrhage Following Tonsillectomy Greater in Adults Than Children?

Tonsillectomy is a commonly performed procedure in the United States, with an estimated 400,000 cases annually. While the majority are performed in the pediatric population, approximately 100,000 (25%) are conducted in adults, based on data from the 2010 National Hospital Ambulatory Medical Care Survey. Although technically straightforward, tonsillectomy carries a risk of uncommon but potentially serious complications, the most significant of which is posttonsillectomy hemorrhage (PTH). In children, PTH rates are well characterized, typically ranging from 1% to 5% [1]. However, PTH incidence in adults is less consistently reported, and in clinical practice, adult patients are often counseled using pediatric data. Multiple studies suggest that PTH rates may be significantly higher in adults [2-5]. Accurate incidence data in adults is therefore essential for effective preoperative counseling and for establishing benchmarks in quality-of-care assessment. This article aims to summarize current literature on PTH incidence in adults and compare it to that in pediatric patients. Pediatric PTH rates have been extensively documented and generally remain low across large-scale studies. Johnson et al. [1] conducted a national retrospective cohort study using data from the Pediatric Health Information System (PHIS), analyzing 96,415 pediatric tonsillectomy cases from 2016 to 2021. PTH, defined by 30-day return visits to the ED or hospital with the ICD-10 diagnosis code for postoperative hemorrhage, was observed in 2.18% of patients. The authors proposed potential quality benchmarks by estimating PTH percentiles at the fifth (1.17%), 50th (1.97%), 95th (4.75%), and 99th (6.39%) percentiles. Notably, patients > 12 years had significantly higher odds of bleeding (OR 2.48), suggesting that older age may influence PTH risk. While pediatric benchmarks such as these are well established, comparable data for adults remain limited. Bhattacharyya et al. [2] examined adult PTH rates in a cross-sectional, multistate analysis using 2010 data from the State Ambulatory Surgery Databases in New York, Florida, Iowa, and California. Among 7748 adult tonsillectomy cases, 11.6% had at least one postoperative revisit within 14 days, with 4.8% presenting with PTH and 2.2% requiring procedural intervention. The study's strength lies in its population-level scope, but the study relied on administrative codes for hemorrhage, which can suffer from misclassification or incomplete data capture. Additionally, the absence of standardized definitions for bleeding severity may have led to underreporting or variability in classifying PTH events. Contemporary classifications of PTH are generally separated into Class I PTH (historical or observed bleeding not requiring any intervention), Class II hemorrhages (bleeding requiring control with local measures), and Class III (requiring intervention in the OR) [3]. A recently published retrospective cohort study by McLean et al. [3] used an institution-specific PTH database involving 1057 adults who underwent tonsillectomy between 2012 and 2019. PTH events were explicitly categorized using the above classification system. The overall PTH rate observed was 11.9% (Class I–III). Additionally, the authors found that the incidence of Class II and III bleeding was 8.5%, a figure notably higher than the expected rate in pediatric cohorts, even when excluding minor Class I bleeds. This study's primary strength was the utilization of a PTH-specific database maintained by the authors' institution, which could explain the higher rate of documentation of PTH, particularly with Class I and Class II bleeds, which are generally not captured in other studies investigating Class III PTH rates only. In a Japanese study by Inuzuka et al. [4], 325 adults undergoing tonsillectomy between 2014 and 2018 were observed in an inpatient setting. The incidence of PTH in this study was 21.8% (Class I–III) and 3.1% needed hemostatic procedures (Class II and III). Routine postoperative inpatient admission for these patients likely allowed for higher yield in detection of PTH events, particularly those that might not prompt emergency visits had they been in an ambulatory setting. This is demonstrated in the much higher proportion of Class I PTH observed in this study, compared to Class II and III rates. This study highlights how the study setting and intensity of surveillance can significantly influence the observed incidence of PTH. Similarly, Ikoma et al. [5] analyzed a mixed cohort of 692 patients, including 369 adults and 323 pediatric patients, all hospitalized postoperatively for at least 8 days. The overall PTH rate for the mixed cohort was 11.6% (Class I–III), but it was observed to be much higher in adults (20.3%). Among the 80 patients with PTH, 93.8% (p < 0.01) were adults, and adults also accounted for 88.9% of Class III bleeds (p < 0.01). Multivariable logistic regression identified adult age as a significant risk factor for hemorrhage (OR 18.9, 95% CI: 7.49–47.7, p < 0.01), underscoring a markedly elevated risk relative to pediatric patients. The limitations of the study include significant heterogeneity within the literature with regards to study design, which leads to variability of reported PTH rates in the adult population. One major challenge in interpreting the literature is the inconsistent definition of what is considered a “clinically significant” PTH event. Future research should prioritize studies from diverse populations that standardize PTH classification and stratify risk by clinical factors such as surgical indication (infectious versus obstructive), patient comorbidities, operative technique (cold steel, bipolar, monopolar cautery), extent of surgery (tonsillotomy versus tonsillectomy), and perioperative medications (e.g., NSAID use). Accurate estimates of the risk of PTH in adults will aid in counseling and enhance shared decision-making for patients considering tonsillectomy. It will also aid the development of adult-specific PTH quality indicators and benchmarks, which will allow for targeted quality improvement initiatives aimed at reducing postoperative morbidity. Despite differences in study design and definitions of hemorrhage, there is a consistent observation across retrospective and cross-sectional studies that adults experience higher rates of clinically significant PTH than children. Whereas reported pediatric bleeding rates are around 1%–5% [1], adult rates are reported to be higher from 5% to over 20% [2-5], particularly when using more sensitive methods of event detection (such as monitoring during inpatient stay [4, 5], or dedicated PTH databases [3]) or inclusive hemorrhage definitions (inclusion of Class I bleed [3-5]). The reported differences between adult and pediatric PTH rates indicate that it is inappropriate to extrapolate pediatric data to the adult population. Counseling of adult patients undergoing tonsillectomy should include discussion of risks of PTH using data from the adult population. The authors declare no conflicts of interest. This manuscript evaluated results of previously-published studies. No new research data was generated.