- Email: [email protected]
Pediatric adenoidectomy in the very young child and indications for postoperative inpatient admission
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Pediatric adenoidectomy in the very young child and indications for postoperative inpatient admission
T
Stephen R. Chorneya,b, Julia F. Daileya, Karen B. Zura,b,∗ a b
Division of Otolaryngology, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, 19104, USA Department of Otorhinolaryngology - Head and Neck Surgery, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, 19104, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: Adenoidectomy Obstructive sleep apnea Outpatient Postoperative complications Sleep disordered breathing Sleep medicine
Objectives: To determine the rate of significant respiratory events following adenoidectomy in young patients and to identify factors that would prompt inpatient admission postoperatively. Methods: A retrospective chart review was performed of consecutive adenoidectomy surgeries at a high-volume, tertiary-care children's hospital between 2016 and 2018. Children under 3.5 years of age who had surgery for obstructive symptoms were included. Patients were grouped by age (youngest ≤1.5 years, middle 1.6–2.5 years, and oldest 2.6–3.5 years). We excluded patients having revision surgery, a concomitant tonsillectomy, or additional major surgical procedure. Results: There were 353 patients that met inclusion criteria. The three age groups were similar with respect to all characteristics except age (p < .001), body mass index (p < .001), and percentage of Black or African American children (p = .02). Patients under 1.5 years more often had preoperative polysomnography (p = .02) with a lower oxygen saturation nadir (p = .04), and were more likely to have surgery for obstructive sleep apnea (p < .001). No differences were found between age groups with respect to recovery room issues, nurse triage calls, or readmissions within 30 days of surgery. An elective admission rate in the cohort was 35.1%, and this was age-group dependent with 79.5% of the youngest group being admitted (p < .001). On admission, 16.9% of all patients had admission events requiring positive pressure support, intensive care unit admission, or prolonged hospitalization, which was similar across all age groups (p = .67). Events were more common in younger patients (17 mos. vs 20 mos., p = .07), those with more comorbidities (74.8% vs 51.5%, p = .06) and significantly higher in those with severe preoperative polysomnogram variables (p < .001). Based on multivariate regression analysis, younger children (OR: 13.7, 95% CI: 6.5 - 29.0, p < .001) or children with an AHI over 5 events/hr (OR: 32.3, 95% CI: 3.4 - 303.2, p = .005) were more likely to have significant events on admission. Conclusions: Significant respiratory events are uncommon after adenoidectomy for obstructive symptoms, even in very young children. However, for children under 1.5 years of age or those with AHI scores above 5 events/hr, postoperative admission for monitoring is recommended. Clinical judgement should be used when considering outpatient surgery for older children or those with comorbidities.
1. Introduction Adenoidectomy is the third most common ambulatory procedure in pediatric otolaryngology, with an estimated 129,000 procedures annually in the United States [1]. With very low postoperative complication rates following adenoid surgery [2–5] short monitoring and same-day discharge is a reasonable approach for most children [6]. The trend towards more judicious use of inpatient surgical services over the last 10–15 years has resulted in 99.4% of adenoidectomies in 2006 being performed on an outpatient basis [7].
In the pediatric population, sleep disordered breathing is most commonly due to tonsillar and adenoid tissue hypertrophy [8]. While the American Academy of Otolaryngology - Head and Neck Surgery (AAO-HNS) defines tonsillectomy as a surgical procedure that often, but not always, includes the removal of the adenoid with the tonsils, the AAO-HNS guideline on pediatric tonsillectomy does not apply to children younger than age 2 or to children having adenoidectomy alone [9,10]. Although performing an adenoidectomy does not address the oropharyngeal obstruction secondary to tonsillar hyperplasia [11], there has never been any level 1 study looking at the efficacy of
∗ Corresponding author. Division of Otolaryngology, Children's Hospital of Philadelphia, Philadelphia, 3401 Civic Center Blvd., 1 Wood ENT, Pennsylvania, 19104, USA. E-mail addresses: [email protected] (S.R. Chorney), [email protected] (J.F. Dailey), [email protected] (K.B. Zur).
https://doi.org/10.1016/j.ijporl.2019.109796 Received 19 September 2019; Received in revised form 22 November 2019; Accepted 22 November 2019 Available online 25 November 2019 0165-5876/ © 2019 Published by Elsevier B.V.
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
Table 1 Patient characteristics adenoidectomies for obstructive symptoms grouped by age at surgery: 2016–2018.
Total No. (%) Mean age at surgery, yr. (SD) Gender, No. (%) Male Female Gestational age, wk. (SD) NICU admission, No. (%) NICU intubation, No. (%) Body mass index, kg/m2 (SD) Race, No. (%) White Black or African American Asian, Indian, Othera Comorbidities, No. (%) Reactive Airway Disease Otherb None
All Patients
≤1.5 Years
1.6–2.5 Years
2.6–3.5 Years
P Value
353 2.2 (0.7)
73 (20.7) 1.2 (0.3)
164 (46.4) 2.0 (0.3)
116 (32.9) 3.0 (0.3)
< .001
236 (66.9) 117 (33.1) 37.6 (3.5) 89 (25.2) 24 (6.8) 16.8 (1.7)
48 (65.8) 25 (34.2) 37.8 (3.2) 20 (27.4) 4 (5.5) 17.2 (1.7)
109 (66.5) 55 (33.5) 37.7 (3.8) 37 (22.6) 10 (6.1) 17.1 (1.6)
79 (68.1) 37 (31.9) 37.5 (3.5) 32 (27.6) 10 (8.6) 16.3 (1.8)
.95
157 (44.5) 139 (39.4) 57 (16.1)
29 (39.7) 36 (49.3) 8 (11.0)
74 (45.1) 69 (42.1) 21 (12.8)
54 (46.6) 34 (29.3) 28 (24.1)
.02
113 (32.0) 60 (17.0) 180 (51.0)
25 (34.2) 13 (17.8) 35 (47.9)
52 (31.7) 20 (12.2) 92 (56.1)
36 (31.0) 27 (23.3) 53 (45.7)
.15
.84 .57 .74 < .001
SD: standard deviation. P values < 0.05 represent a statistically significant result. a Other race includes small numbers of patients who refused to identify on demographics. b Other comorbidities: Neuromuscular Disorder, Cardiopulmonary Disease, Hematological Disorder, Craniofacial Syndrome, and Trisomy 21.
tube placement, ear exam under anesthesia, cerumen removal, or tympanostomy tube removal with or without paper patch) or minor otolaryngology procedure (nasal cautery, lingual frenotomy, or skin tag excision). Records were analyzed for patient demographics including date of birth, gestational age, gender, body mass index, race, history of NICU admission, and history of NICU intubation. Comorbidities were recorded and grouped according to category. The clinical work-up included whether a lateral X-ray or a nasopharyngoscopy was obtained and these were included along with the reported size of the adenoid. Preoperative polysomnography (PSG), if available, was recorded with Apnea-Hypopnea Index, nadir Oxygen saturation, and peak end tidal CO2 levels. Operative reports were assessed for date of procedure, surgical procedure, surgical indication, intraoperative complications, adenoidectomy technique, and size of adenoid tissue. Notes from physician and nursing staff were included in order to identify any perioperative events or additional support in the post anesthesia care unit (PACU), floor, or intensive care unit. Progress notes and vitals were included to assess for respiratory and non-respiratory events. Interventions, if needed, were also recorded from the electronic progress notes. Location of admission (floor vs. intensive care unit) as well as duration of hospitalization were noted. Upon discharge, if a patient's family contacted our division's nursing answering service, this was documented up to 30 days. The nature of the call was recorded if it was related to the adenoid procedure and not the minor otologic or otolaryngological procedures. Readmissions during the first 30 days were also included, along with the indication, outcomes, and disposition of the patient. Data was analyzed using SPSS® Statistics for Windows (Version 25.0 Armonk, NY: IBM Corp.). Continuous variables were expressed in means with standard deviation. Categorical values were represented by the absolute number along with percentage. Tests for statistical significance were performed by using student's t-testing and analysis of variance (ANOVA) testing where appropriate. Fisher's exact testing was utilized for categorical values. Finally, multinomial logistical regression analyses were performed for various risk factors and the influence they had on significant admission events. Regression outputs were given in odds ratios along with 95% confidence intervals. Significance levels were set at p < 0.05.
adenoidectomy alone for the treatment of obstructive sleep apnea (OSA) or sleep disordered breathing (SDB), which may be a reasonable option for select children [12]. In 1996, the AAO-HNS recommended outpatient tonsillectomy for healthy children with an American Society of Anesthesiologists classification I or II, without evidence of obstructive sleep apnea-hypopnea syndrome (OSAHS), and age older than 3 years [13]. Surgeons must be concerned with patients who are at higher risk for respiratory complications postoperatively. These complications could include desaturations secondary to increased swelling and secretions, apneic events, or pulmonary edema [14]. The question remains whether children should be monitored after adenoidectomy for obstructive symptoms similarly to children who undergo adenotonsillectomy (T&A). Adenoid surgery is shorter in duration, does not involve the oropharynx, and is managed without narcotic pain medication at our institution. The main objective of this study is to characterize complications following adenoidectomy without tonsillectomy for obstructive symptoms in young children. Secondary goals are to identify which children benefit from inpatient admission following surgery in order to monitor for respiratory events. This large retrospective review will also provide meaningful data on the preoperative work-up, the adenoid technique, as well as the frequency of nurse triage calls and readmissions following adenoidectomy within the first 30 days. 2. Methods A retrospective chart review was approved with exemption by the Institutional Review Board at Children's Hospital of Philadelphia. Patients under the age of 3.5 years at the time of surgery were included based on Current Procedural Terminology (CPT) code 42830 (adenoidectomy, primary; under age 12). Only patients with obstructive symptoms were included. Patients who underwent an adenoidectomy for recurrent ear infections, chronic adenoiditis, or otitis without obstructive symptoms were excluded. Consecutive electronic medical records of children who had surgery at our main hospital operating room between January 1, 2016 and July 1, 2018 were reviewed. If a patient had a previous adenoidectomy, a concomitant tonsillectomy, an airway intervention such as bronchoscopy, or an additional major surgical procedure, they were excluded from our analysis. Patients who were tracheostomy or ventilatory dependent were not included. An additional minor procedure with adenoidectomy was acceptable, such as a minor otologic procedure (bilateral myringotomy with tympanostomy 2
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
3. Results
middle age group had a suction cautery adenoidectomy and 52.6% of the oldest group had a microdebrider adenoidectomy (p < .001) (Table 3). There were no recorded intraoperative complications. Events during the perioperative period and the postoperative hospitalization were characterized as minor or significant and were overwhelmingly respiratory-related. Issues in the PACU after adenoidectomy surgery were uncommon, occurring in 8.2% of patients (3.4% significant, 4.8% minor). No differences were seen in these events between age groups (p = .41). There were three episodes of bronchospasm (0.9%), two in the youngest and one in the middle age group, and each was managed with albuterol. There was no laryngospasm and no patient required reintubation. Overall, the planned admission rate was 35.1%, which was age-group dependent as 79.5% of the youngest patients were admitted (p < .001). On admission, 16.9% of patients had a significant event, and this was similar across all age groups (p = .67). Nurse triage calls (27.8%, p = .97), emergency department visits (4.0%), and readmissions within 30 days (0.8%) were not significantly different between groups (p = .51) (Table 4). Not shown was one minor bleed in the PACU that self-resolved and four nurse calls for bleeds (three in the middle group and one in the oldest), with one requiring inpatient readmission for 2 days after bedside packing placement (middle age group). No bleeding occurred in patients with bleeding disorders or with nasal cautery procedures. On readmission, three patients were admitted longer than 24 hours. The most serious complication occurred in a 14 month-old, full term child with reactive airway disease who was readmitted with pneumonia for 8 days requiring video-assisted thoracoscopic surgery (VATS) with chest tube placement for pleural effusion and empyema. Poor oral intake was not an issue for any children in the immediate postoperative period, however we did have four children requiring admission to the ED for hydration and pain control for inadequate oral intake within 30 days. There were no cases of Grisel's syndrome in our cohort. When we looked at characteristics of patients scheduled to be admitted after adenoidectomy, they tended to be younger (1.6 yrs. vs. 2.4 yrs, p < .001), have a higher body mass index (17.2 kg/m2 vs 16.6 kg/m2, p = .01), and were more likely to be Black or African American (49.2% vs 34.1%, p = .001). Additionally, admitted patients with a preoperative sleep study had a higher mean AHI, lower mean nadir O2 saturation, and higher peak end-tidal CO2 (Table 5). Patients with more eventful admissions (i.e. placement of nasopharyngeal airway, need for positive pressure, admission to the ICU, or more prolonged admission) were younger (1.4 yrs vs 1.7 yrs, p = .07), had more comorbidities (74.8% vs 51.5%, p = .06), were more likely to be White (61.9% vs 39.8%, p = .001), and had more severe grouping of AHI, O2 nadir, and end tidal CO2 on PSG (p < .001). Interestingly, while age at surgery was lower, history of NICU admission was higher, and presence of comorbidities were higher in the patients with complicated admissions, these failed to meet statistical significance (Table 6). Subgroup analysis was performed on the youngest group, focusing on events in children with and without a preoperative polysomnogram
There were 353 patients between January 1, 2016 and July 1, 2018 who met inclusion criteria. Table 1 shows the patient characteristics of the study population as well as the characteristics of patients in our three age groups: youngest (≤1.5 years), middle (1.6–2.5 years), and oldest (2.6–3.5 years). The average patient was 2.2 (SD: 0.7) years of age at surgery, with a mean gestational age of 37.6 (SD: 3.5) weeks. Our population included 49.0% of patients diagnosed with at least one comorbidity, with reactive airway disease (32.0%) representing the majority. Not listed, but grouped into other comorbidities, were: Neuromuscular Disorder (4.8%), Cardiopulmonary Disease (4.0%), Hematological Disorder (3.7%), Craniofacial Syndrome (2.0%), and Trisomy 21 (1.7%). Our three population groups were statistically similar with respect to all characteristics except age (p < .001), body mass index (p < .001), and race, with a higher frequency of Black or African American children in the younger group (49.3%, p = .02). The majority of our patients had a preoperative lateral neck X-ray showing moderate or fully obstructing adenoids in 87.7% of children. Nasopharyngoscopy was obtained less frequently for 11.0% of patients, with 89.7% having moderate or fully obstructing adenoids. In our group, 21.5% of children had a preoperative PSG, with a mean AHI of 9.8 (SD: 15.9), mean nadir oxygen saturation of 86.8% (SD: 5.0), and mean peak end tidal CO2 of 52.0 Torr (SD: 5.6) (Table 2). When comparing age groups, we found that nasopharyngoscopy was performed more commonly in the youngest group (p < .001), and the middle group had a higher ratio of non-obstructive and mildly obstructive adenoids on nasopharyngoscopy (p < .001). Additionally, a larger percentage (34.2%) of the youngest age group obtained a preoperative PSG (p = .02) and had a lower mean Oxygen saturation nadir (84.7%, p = .04) than the other two groups. Surgical indications in our study population were adenoid hypertrophy (92.1%), nasal obstruction (46.5%), sleep disordered breathing (46.5%), snoring (24.6%), and obstructive sleep apnea (16.4%) (Table 3). Indications were similar across age groups except for obstructive sleep apnea, which was performed for 32.9% of the youngest group, compared to 11.6% and 12.9% (p < .001). The majority of our patients (66.9%) underwent adenoidectomy with a minor otologic surgery, namely bilateral myringotomy with tympanostomy tube placement. Adenoidectomy alone was performed in 31.2% of cases and adenoidectomy with a minor otolaryngological procedure performed in 2.0% of cases. Older patients were less likely to have a minor otologic procedure and more likely to have an adenoidectomy alone (p = .01). Moderate or fully obstructing adenoids were found intraoperatively in 97.5% of our patients. The three surgical techniques used in our institution were microdebrider (37.4%), suction cautery (36.0%), and coblation (26.6%). The choice of technique is surgeon-dependent, with 14 surgeons in our group during the study period. Significantly more coblation adenoidectomy occurred in the youngest patients (38.4%), while 47.0% of the Table 2 Clinical evaluation adenoidectomies grouped by age at surgery: 2016–2018.
Lateral Neck X-ray, No. (%) Non/Mildly Obstructive Moderate/Fully Obstructive Nasopharyngoscopy, No. (%) Non/Mildly Obstructive Moderate/Fully Obstructive Polysomnography, No. (%) Apnea Hypopnea Index, Mean AHI (SD) Nadir Oxygen Saturation, Mean % (SD) Peak End-Tidal CO2, Mean Torr (SD)
All Patients
≤1.5 Years
1.6–2.5 Years
2.6–3.5 Years
P Value
212 (60.1) 26 (12.3) 186 (87.7) 39 (11.0) 4 (10.3) 35 (89.7) 76 (21.5) 9.8 (15.9) 86.8 (5.0) 52.0 (5.6)
36 (49.3) 4 (11.1) 32 (88.8) 22 (30.1) 2 (9.1) 20 (90.9) 25 (34.2) 12.7 (12.1) 84.7 (5.1) 52.9 (6.6)
102 (62.2) 14 (13.7) 88 (86.3) 8 (4.9) 2 (25.0) 6 (75.0) 29 (17.7) 9.1 (21.9) 87.9 (4.7) 52.0 (5.0)
74 (63.8) 8 (10.8) 66 (89.2) 9 (7.8) 0 (0) 9 (100) 22 (19.0) 7.5 (8.8) 87.9 (4.9) 51.1 (5.2)
.11 .32
SD: standard deviation. P values < 0.05 represent a statistically significant result. 3
< .001 < .001 .02 .52 .04 .58
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
Table 3 Surgical data adenoidectomies grouped by age at surgery: 2016–2018. All Patients
≤1.5 Years
1.6–2.5 Years
2.6–3.5 Years
P Value
325 (92.1) 164 (46.5) 164 (46.5) 87 (24.6) 58 (16.4)
66 25 34 17 24
(90.4) (34.2) (46.6) (23.3) (32.9)
151 (92.1) 82 (50.0) 84 (51.2) 39 (23.8) 19 (11.6)
108 (93.1) 57 (49.1) 46 (39.7) 31 (26.7) 15 (12.9)
.82 .06 .16 .83 < .001
236 (66.9) 110 (31.2) 7 (2.0)
51 (69.9) 20 (27.4) 2 (2.7)
118 (72.0) 46 (28.0) 0 (0)
67 (57.8) 44 (37.9) 5 (4.3)
.01
9 (2.5) 344 (97.5)
2 (2.7) 71 (97.3)
5 (3.0) 159 (97.0)
2 (1.7) 114 (98.3)
.83
132 (37.4) 127 (36.0) 94 (26.6)
26 (35.6) 19 (26.0) 28 (38.4)
45 (27.4) 77 (47.0) 42 (25.6)
61 (52.6) 31 (26.7) 24 (20.7)
< .001
a
Surgical Indication , No. (%) Adenoid Hypertrophy Nasal Obstruction Sleep Disordered Breathing Snoring Obstructive Sleep Apnea Surgical Procedure, No. (%) Adenoidectomy & minor ear surgeryb Adenoidectomy only Adenoidectomy & minor ENT surgeryc Description of Adenoids, No. (%) Non/Mildly Obstructive Moderate/Fully Obstructive Adenoidectomy Technique, No. (%) Microdebrider Suction Cautery Coblation
P values < 0.05 represent a statistically significant result. a Patients may have multiple surgical indications. b Minor otologic surgery: bilateral myringotomy with tympanostomy tube placement, ear exam under anesthesia, cerumen removal, or tympanostomy tube removal with or without paper patch. c Minor ENT surgery: nasal cautery, lingual frenotomy, or skin tag excision. Table 4 Postoperative experience adenoidectomies grouped by age at surgery: 2016–2018.
Issues in PACU, No. (%) Significant respiratory eventa Minor respiratory eventb Admission Rate (planned), No. (%) Significant admission event, No. (%)c Nurse Triage Calls, No. (%) Fever, URI Pain, poor oral intake Bleeding Otherd Readmission Within 30 Days, No. (%) Inpatient readmissione Emergency department observationf
All Patients
≤1.5 Years
1.6–2.5 Years
2.6–3.5 Years
P Value
12 (3.4) 17 (4.8) 124 (35.1) 21 (16.9)
1 (1.4) 6 (8.2) 58 (79.5) 12 (20.7)
6 (3.7) 8 (4.9) 51 (31.1) 7 (13.7)
5 (4.3) 3 (2.6) 15 (12.9) 2 (13.3)
.41
74 (21.0) 18 (5.1) 4 (1.1) 2 (0.6)
16 (21.9) 3 (4.1) 0 1 (1.4)
34 (20.7) 8 (4.9) 3 (1.8) 1 (0.6)
24 (20.7) 7 (6.0) 1 (0.9) 0
.97
3 (0.8) 14 (4.0)
1 (1.4) 2 (2.7)
2 (1.2) 9 (5.5)
0 3 (2.6)
.51
< .001 .67
URI: Upper respiratory tract infection. P values < 0.05 represent a statistically significant result. a Significant airway event: Bronchospasm, oxygen desaturation below 90%, use of high-flow nasal cannula, bag-mask ventilation, use of Continuous Positive Airway Pressure (CPAP), and/or placement of nasopharyngeal airway. b Minor airway event: Oxygen desaturation to 90% with or without placement of low flow oxygen delivery such as nasal cannula, facemask, or blow-by. c Significant admission event: Oxygen desaturation below 85% requiring: use of high flow nasal cannula, bag-mask ventilation, use of Continuous Positive Airway Pressure (CPAP), and/or placement of nasopharyngeal airway; admission to the Intensive Care Unit (ICU) or admission beyond 24 h after surgery. d Other triage calls include: nausea, vomiting, constipation, diarrhea, or neck pain. e Inpatient readmission: significant events that required additional interventions, admission to the floor, or admission beyond 24 h. f Emergency Department observation include patients with pain, nausea, fevers, or poor PO who were monitored for less than 24 h for analgesia and intravenous hydration.
chronic or recurrent otitis media, chronic or recurrent sinusitis, and upper airway obstruction, which includes nasal obstruction, mouth breathing, or snoring [15]. Performing an adenoidectomy alone is a reasonable option for obstructive symptoms such as sleep disordered breathing or obstructive sleep apnea in select children. A survey of AAO-HNS members from New York and the American Society of Pediatric Otolaryngology (ASPO) determined that most surgeons would perform an adenoidectomy alone in a child with OSA and enlarged adenoids, but with small tonsils [16]. Adenoidectomy mitigates early postoperative complications following adenotonsillectomy, including respiratory compromise, hemorrhage, pain, fever, nausea and vomiting, and dehydration [17]. The AAO-HNS clinical practice guideline recommends T&A for OSA [9,10]. Consequently, much of the literature does not separate tonsil and adenoid surgery with respect to complications and the role various risk factors have on respiratory events. We
(Table 7). In children under 1.5 years, age at surgery (p = .81), gender (p = .14), gestational age (p = .65), BMI (p = .57), and presence of comorbidities (p = .41) were similar between those who did and did not have a PSG. While we did admit children more often with a preoperative PSG in this age group (92.0% vs 72.9%, p = .05), rates and severity of events during admission were similar. Finally, multivariate logistical analysis was performed looking at risk factors for children who had significant events on admission (Table 8). Among admitted children, the development of an event was predicted by younger age at surgery (OR: 13.7, 95% CI: 6.5–29.0, p < .001) or an AHI above 5 events/hr (OR: 32.3, 95% CI: 3.4–303.2, p = .005). 4. Discussion Adenoidectomy without tonsillectomy is indicated for children with 4
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
management for patients who are obese, have complicated medical histories, have severe OSA, or are younger than 3 years [9,18]. Patients younger than 3 years and those with multiple comorbidities should be monitored in the hospital because they have relatively high rates of airway complications [19,20]. One group found that children under 3 years of age have a two-fold risk of respiratory complications compared to children 3–6 years [21]. Consequently, admission rates following T& A at pediatric hospitals range from over 94% for children under 2 years with OSA and at least one medical comorbidity, to 14% for children older than 5 years without OSA or comorbidities. However, there still remains significant practice variability even after controlling for age, comorbidities, and OSA [22]. Based on our results, children under 1.5 years of age who have an adenoidectomy alone for obstructive symptoms should be monitored as an inpatient. We did not identify any significant difference between our three age groups with respect to immediate issues in the PACU, frequency of nurse triage calls, or readmissions within 30 days after surgery. Nonrespiratory events such as nausea, vomiting, or poor oral intake were extremely rare. From a bleeding standpoint, there was one self-limited bleed in the PACU immediately after surgery that resolved with topical oxymetazoline. Within the first week, there were four triage calls for bleeding, with one requiring a 48-hour inpatient readmission after placing non-absorbable packing in the emergency department. The rate of bleeding post adenoidectomy has been previously reported in the literature with rates ranging from 0.22 to 1% [2,4,5]. Since our study looks at a subset of patients with obstructive symptoms and does not reflect our entire patient population who undergo adenoidectomy, we cannot extrapolate the global rate of adenoidectomy bleeding at this institution. Overall, our patients had very few significant respiratory-related events. Of the events we recorded, nearly all of them responded to simple interventions. There were three episodes of bronchospasm (0.9%), two in the youngest and one in the middle age group, but these were managed conservatively with albuterol and supplemental oxygen. There was no laryngospasm and no patient required reintubation. The adenotonsillectomy literature has identified rates of 1.6% for laryngospasm and 1.8% for bronchospasm, with bronchospasm occurring after faster surgeries in patients with underlying asthma [23]. Thirty-two percent of our population had a diagnosis of reactive airway disease and this did not impact respiratory event rates during the recovery period. Postoperative respiratory complications can occur in up to 25% of children with OSA after adenotonsillectomy, including oxygen desaturation, atelectasis, pneumonia, pulmonary edema, pleural effusion, pneumothorax, pneumomediastinum, or inspiratory stridor with increased work of breathing [21]. In a large study, children between 1 and 2 years old with a history of nasal obstruction from adenoid hypertrophy, gastroesophageal reflux disease, prematurity, and/or cardiovascular anomalies were at risk for early complications warranting closer observation [24]. Race, gestational age, and BMI were not correlated with complicated events. Further, all of our surgeries were performed for obstructive symptoms among many medically complex patients. While there was a trend for more comorbidities among patients with significant events, this was not statistically significant (p = .06). Therefore, children with comorbidities should raise concern for possible inpatient complications, but our data did not find a clear correlation for this following adenoidectomy. Polysomnogram severity appeared to predict need for support during the inpatient admission following adenoid surgery among the 21.5% of our patients that obtained a preoperative PSG. We found more events in children with an AHI over 5 events per hour, O2 nadir less than 85%, or peak end tidal CO2 over 50 Torr on preoperative PSG (p < .001). Logistical regression analysis identified patients with an AHI over 5 events per hour as having 31 times greater likelihood of a significant respiratory event. While O2 nadir and peak end tidal CO2 did have the same impact on regression modeling, the rates of complicating events in the higher groups suggest that these children warrant close
Table 5 Admission characteristics adenoidectomies grouped by age at surgery: 2016–2018.
Total, No. (%) Mean age at surgery, yr. (SD) Gender, No. (%) Male Female Gestational age, wk. (SD) Body mass index, kg/m2 (SD) Race, No. (%) White Black or African American Asian, Indian, Othera Comorbidities, No. (%) Reactive Airway Disease Otherb None Sleep Study Parameters Apnea Hypopnea Index, Mean AHI (SD) Less than or equal to 5, No. (%)c Greater than 5, No. (%) Nadir Oxygen Saturation, Mean % (SD) Over 85%, No. (%) Less than or equal to 85%, No. (%) Peak End-Tidal CO2, Mean Torr (SD) Less than 50 Torr, No. (%) Greater or equal to 50 Torr, No. (%)
Planned Admission
Outpatient Surgery
P Value
124 (35.1) 1.6 (0.6)
229 (64.9) 2.4 (0.6)
< .001
83 (66.9) 41 (33.1) 37.4 (3.4) 17.2 (1.8)
153 (66.8) 76 (33.2) 37.7 (3.6) 16.6 (1.7)
.54
54 (43.5) 61 (49.2) 9 (7.3)
103 (45.0) 78 (34.1) 48 (21.0)
.001
40 (32.3) 28 (22.6) 56 (45.2)
73 (31.9) 32 (14.0) 124 (54.1)
.09
14.4 (19.4)
3.4 (3.7)
.002
19 (15.3)
28 (12.2)
< .001
25 (20.2) 85.7 (5.3)
4 (1.7) 88.3 (4.3)
.02
27 (21.8) 17 (13.7)
26 (11.4) 6 (2.6)
< .001
53.2 (6.0)
50.5 (4.7)
.04
13 (10.5) 28 (22.6)
14 (6.1) 18 (7.9)
< .001
.44 .01
SD: standard deviation. P values < 0.05 represent a statistically significant result. a Other race includes small numbers of patients who refused to identify on demographics. b Other comorbidities: Neuromuscular Disorder, Cardiopulmonary Disease, Hematological Disorder, Craniofacial Syndrome, and Trisomy 21. c Four children had an Apnea-Hypopnea Index (AHI) less than 1 event/hr (normal): one in the middle group and three in the older age group.
focus on adenoid-only surgery in patients with many of the same risk factors and surgical indications. Adenoidectomy is a milder intervention than adenotonsillectomy, with minimal risks of oropharyngeal swelling, rare bleeding, shorter anesthetic time, and postoperative pain managed without narcotics. Therefore, we suspect that physicians caring for these children could be more aggressive with discharge practices and potentially perform fewer adenoidectomies in an inpatient operating suite. Our institution currently requires children under 2.5 years of age to have their adenoidectomy performed in the main hospital operating room and not at an ambulatory surgical center. The indication for admission post-adenoidectomy is mostly surgeondependent and based on associated comorbidities or severity of sleep apnea. In our large cohort, the post-adenoidectomy elective admission rate of 35.1% was age-group dependent, with 79.5% of the youngest group being admitted (p < .001). Despite balanced event rates across all three groups (20.7%, 13.7%, 13.3%, p = .67), we found that patients with the most significant events were just under 17 months old compared to around 20 months (p = .07). On multivariate regression analysis, younger patients were 13 times more likely to develop a postoperative respiratory event. Age has historically been an important variable in the decision to admit children after adenotonsillectomy. The American Academy of Pediatrics (AAP) and AAO-HNS suggest similar risk factors for postoperative complications and recommend inpatient
5
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
Table 6 Significant admission events adenoidectomies grouped by age at surgery: 2016–2018.
Total, No. (%) Mean age at surgery, yr. (SD) Gender, No. (%) Male Female Gestational age, wk. (SD) Body mass index, kg/m2 (SD) Race, No. (%) White Black or African American Asian, Indian, Othera Comorbidities, No. (%) Reactive Airway Disease Otherb None Sleep Study Parameters Apnea Hypopnea Index, Mean AHI (SD) Less than or equal to 5, No. (%) Greater than 5, No. (%) Nadir Oxygen Saturation, Mean % (SD) Over 85%, No. (%) Less than or equal to 85%, No. (%) Peak End-Tidal CO2, Mean Torr (SD) Less than 50 Torr, No. (%) Greater or equal to 50 Torr, No. (%)
Admissions Without Significant Events
Admissions With Significant Events
P Value
103 (83.1) 1.7 (0.6)
21 (16.9) 1.4 (0.5)
.07
70 (68.0) 33 (32.0) 37.5 (3.4) 17.2 (1.9)
13 (61.9) 8 (38.1) 37.1 (3.4) 17.1 (1.7)
.83
41 (39.8) 54 (52.4) 41 (39.8)
13 (61.9) 7 (33.3) 1 (4.8)
.001
33 (32.0) 20 (19.4) 50 (48.5)
7 (33.3) 8 (38.1) 6 (28.6)
.06
12.2 (16.3) 15 (14.6) 16 (15.5) 86.4 (4.4) 21 (20.4) 10 (9.7) 53.1 (5.8) 10 (9.7) 20 (19.4)
19.8 (25.4) 4 (19.0) 9 (42.9) 84.2 (6.9) 6 (28.6) 7 (33.3) 53.6 (6.7) 3 (14.3) 8 (38.1)
.24 < .001
.74 .81
.20 < .001 .79 < .001
SD: standard deviation. P values < 0.05 represent a statistically significant result. a Other race includes small numbers of patients who refused to identify on demographics. b Other comorbidities: Neuromuscular Disorder, Cardiopulmonary Disease, Hematological Disorder, Craniofacial Syndrome, and Trisomy 21.
admission was to the ICU, with 80% of these in the under 1.5-year-old age group. Of note, 7 of 10 in the youngest group were discharged from the ICU on postoperative day (POD) 1, while the other 3 were discharged in 48–72 hours. This calls into question the role of ICU admission afterwards in adenoidectomy patients. A previous study has suggested that in children under the age of 12 months who have an adenoidectomy for OSA, postoperative monitoring in the intensive care unit for 24 hours is highly recommended [27]. Depending on staffing and institutional comfort, many of these children could be managed on the floor with close nursing care as most of our ICU patients did not
attention as well. To date, the relationship between PSG variables and adenoidectomy complications have not been well-studied. Looking at adenotonsillectomy outcomes, one group noted that patients with desaturations following T&A requiring supplemental oxygen had significantly higher AHI, (31.8 vs 14.1; p = .001), higher hypopnea index (22.6 vs 8.9; p = .004), and lower nadir oxygen saturation (72% vs 84%; p = .001) [25]. Further, O2 nadir less than 80% or peak CO2 greater than 60 mm Hg on PSG are strong and consistent predictors of postoperative respiratory complications following adenotonsillectomy in children with OSA [26]. For 12 of our patients, postoperative
Table 7 Children under 1.5 years comparing PSG without PSG adenoidectomies: 2016–2018.
Total, No. (%) Mean age at surgery, yr. (SD) Male gender, No. (%) Gestational age, wk. (SD) Body mass index, kg/m2 (SD) Comorbidities, No. (%) Admitted, No. (%) PACU Events, No. (%) PACU Event Details, (No., %)
Admission Events, No. (%) Admission Event Details, (No., %)
No Sleep Study
Preoperative Sleep Study
48 (65.8) 1.2 (0.3) 29 (60.4) 37.9 (3.1) 17.1 (1.7) 24 (50.0%) 35 (72.9) 5 (10.4) · Broncospasm, albuterol given, admitted to floor (1, 2.1%) · O2 NC, weaned on floor (1, 2.1%) · Desats requiring NC or HFNC, with ICU admission (3, 6.3%) 5 (14.2) · ICU, desats to low 70's needing 1L NC, weaned (2, 5.7%) · ICU, NP airway and HFNC for desats to low 80's (1, 2.9%) · Floor, 2L NC weaned for desats to upper 70's-low 80's (2, 5.7%) · Readmit for 8 days for pneumonia, VATS for empyema (1, 2.1%)
25 (34.2) 1.2 (0.3) 19 (76.0) 37.9 (3.4) 17.4 (1.6) 14 (56.0%) 23 (92.0) 2 (8.0) · Bronchospasm in OR, albuterol given, observed in ICU without O2 (1, 4.0%) · Brief desats, 1L NC, weaned on floor (1, 4.0%)
7 (30.4) · ICU, 2L NC weaned to RA (1, 4.3%) · ICU, desats to 82% repositioning, no O2 needed (1, 4.3%) · ICU,no desats or O2 (5, 21.7%)
P value
.81 .14 .65 .57 .41 .05 1.00
.13
ICU = intensive care unit, NC = nasal cannula, NP = nasopharyngeal, HFNC = high flow nasal cannula, desats = desaturations, VATS= Video-Assisted Thoracoscopic Surgery, RA = room air. 6
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
missed in other children who met ambulatory surgery center criteria. Third, only children with obstructive symptoms were included, which was strategic from a study design as we made the assumption that children with obstructive symptoms may have more difficulty with recovery compared to the patients with no pre-existing sleep issues. However, this could lead to an inability to generalize to all children who have an adenoidectomy. To analyze whether we can liberalize the recommendation for admission in the middle or youngest age groups, we will be studying patients who had an adenoidectomy for infection/ adenoiditis. Fourth, grouping various comorbidities due to small numbers limits generalizability and analysis, which may be an important area of post-operative complications. Finally, the ability to determine respiratory complications and measurements of events are limited by the accuracy of the documentation of clinical staff. As such, there could be events that are missed if they were not recorded in the patient's electronic record.
Table 8 Multivariate analysis of risks for significant admission events adenoidectomies: 2016–2018.
Younger age at surgery Gender Younger gestational age Higher body mass index Race Presence of comorbidities Higher Apnea Hypopnea Index Lower Nadir Oxygen Saturation Higher Peak End-Tidal CO2
Odds Ratio
95% Confidence Interval
P Value
13.7 1.1 1.0 0.9 3.5 1.5 32.3 0.6 1.8
(6.5–29.0) (0.5–2.4) (0.9–1.2) (0.7–1.1) (0.7–17.1) (0.6–3.7) (3.4–303.2) (0.1–3.9) (0.3–10.0)
< .001 .39 .81 .40 .09 .25 .005 .68 .46
P values < 0.05 represent a statistically significant result.
require intensive services aside from two patients requiring CPAP overnight. The vast majority of our children had a lateral neck X-ray (60.1%) but a flexible nasopharyngoscopy was performed for 11% of our patients. Notably, our youngest age group was more likely to have a nasopharyngoscopy (30.1%, p < .001). The reasons for this are unclear, but may be related to practitioner preference, whether X-ray studies were previously done by referring clinicians, ease of examination, parental choice (both financial and related to their concern about performing a nasopharyngoscopy on their child), and less radiation. There are inherent issues with measuring size of adenoids, but in our institution, we rely heavily on the lateral neck film. A recent best practice review stated that flexible nasal endoscopy is well tolerated in most children, allows for direct visualization of the adenoid, and is comparable to lateral neck X-ray in cost. Nasal endoscopy also avoids variation due to poor patient position and radiation exposure of lateral neck films. The authors recommend a flexible nasal endoscopy as the best initial choice for evaluation of adenoid size when the tonsils are small, with a lateral neck X-ray reserved for children who are unable to cooperate with a flexible nasal endoscopic exam [28]. Regardless of assessment technique, the decision to perform adenoidectomy alone in our population was nearly universally a result of small tonsils on exam with evidence of large adenoids from X-ray or nasal endoscopy. Microdebrider and suction cautery were the two most common surgical techniques, with a slightly smaller amount of coblation adenoidectomy. We did find some age-group differences, with 38.4% of the youngest patients having coblation adenoidectomy, 47.0% of the middle age group having suction cautery adenoidectomy, and 52.6% of the oldest group having a microdebrider adenoidectomy (p < .001). This difference likely represents practitioner preferences as opposed to a meaningful relationship to the age of the patient. Nonetheless, each technique is used routinely in all age groups. A review of adenoidectomy techniques found that electrocautery adenoidectomy has an advantage with regards to minimizing operative time and blood loss at a lower cost than power-assisted adenoidectomy [29]. The merits and drawbacks to each approach are beyond the scope of this paper, but we did not identify any intraoperative or postoperative complications and therefore find a role for all three techniques. There are several limitations to the results of this study. First, the retrospective nature of the data introduces selection bias particularly with respect to younger children having more documented OSA on PSG. Reluctance to perform tonsillectomy in young patients for risk of severe postoperative complications may have been a reason for clinicians ordering preoperative PSG more frequently. As a result, older children may have similar polysomnogram severity, but do not appear this way given a lower frequency of PSG ordering. Second, we only looked at patients who had surgery at the main operating room and not in our outpatient surgery centers. The tendency to electively admit younger patients after adenoid surgery in our institution may have overestimated the rate of complications in this group, which we may have
5. Conclusions Adenoidectomy without tonsillectomy has a role in the management of select children with upper airway obstruction. Although postoperative respiratory events are rare, we identified age and severity of PSG variables as predictive of events that would warrant admission. In particular, for children under 1.5 years or those with severe PSG variables (AHI above 5 events/hr, nadir O2 below 85%, or peak end tidal CO2 above 55 Torr), inpatient monitoring for respiratory complications is recommended. However, since only a fraction of the patients in our cohort had a preoperative sleep study, we recommend that clinical judgement should be used when considering outpatient surgery in the youngest age group. Furthermore, for patients who are older than 1.5 years of age, consideration can be made for performing an adenoidectomy at a surgical center setting barring additional comorbidities that would exclude their eligibility. Our results provide one of the largest single-center cohorts published on events following adenoidectomy for obstructive symptoms. Future studies should continue to define patient characteristics that determine appropriate time for discharge after adenoidectomy in young children. Declaration of competing interest The authors have no conflicts of interest to report. Acknowledgements The authors would like to acknowledge Terri Giordano DNP, CRNP, CORLN who provided invaluable support during the design and Institutional Review Board submission of this research. References [1] N. Bhattacharyya, Ambulatory pediatric otolaryngologic procedures in the United States: characteristics and perioperative safety, The Laryngoscope 120 (4) (2010 Apr) 821–825. [2] J.P. Windfuhr, Y.S. Chen, Post-tonsillectomy and-adenoidectomy hemorrhage in nonselected patients, Ann. Otol. Rhinol. Laryngol. 112 (1) (2003 Jan) 63–70. [3] A. Hunt, M. Karela, P.J. Robb, Day-case adenoidectomy: outcomes are improved using suction coagulation and prophylactic anti-emetic treatment, Int. J. Pediatr. Otorhinolaryngol. 69 (12) (2005 Dec 1) 1629–1633. [4] J.P. Windfuhr, Y.S. Chen, S. Remmert, Hemorrhage following tonsillectomy and adenoidectomy in 15,218 patients, Otolaryngology-Head Neck Surg. (Tokyo) 132 (2) (2005 Feb) 281–286. [5] K. Thomas, D. Boeger, J. Buentzel, D. Esser, K. Hoffmann, P. Jecker, A. Mueller, G. Radtke, K. Geißler, M. Finkensieper, O. Guntinas-Lichius, Pediatric adenoidectomy: a population-based regional study on epidemiology and outcome, Int. J. Pediatr. Otorhinolaryngol. 77 (10) (2013 Oct 1) 1716–1720. [6] S. Habib, K. Solan, P. Das, Reducing observation time in children post‐adenoidectomy, Clin. Otolaryngol. 44 (2) (2019 Mar) 204. [7] N. Bhattacharyya, H.W. Lin, Changes and consistencies in the epidemiology of pediatric adenotonsillar surgery, 1996–2006, Otolaryngology-Head Neck Surg. (Tokyo) 143 (5) (2010 Nov) 680–684.
7
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
syndrome, Pediatrics 109 (4) (2002) 704–712. [19] M. Amoils, K.W. Chang, O. Saynina, P.H. Wise, A. Honkanen, Postoperative complications in pediatric tonsillectomy and adenoidectomy in ambulatory vs inpatient settings, JAMA Otolaryngology-Head Neck Surg. 142 (4) (2016 Apr 1) 344–350. [20] C.M. Lawlor, C.A. Riley, J.M. Carter, K.H. Rodriguez, Association between age and weight as risk factors for complication after tonsillectomy in healthy children, JAMA Otolaryngology-Head Neck Surg. 144 (5) (2018 May 1) 399–405. [21] M.M. Statham, R.G. Elluru, R. Buncher, M. Kalra, Adenotonsillectomy for obstructive sleep apnea syndrome in young children: prevalence of pulmonary complications, Arch. Otolaryngol. Head Neck Surg. 132 (5) (2006 May 1) 476–480. [22] S.S. Goyal, R. Shah, D.W. Roberson, M.L. Schwartz, Variation in post‐adenotonsillectomy admission practices in 24 pediatric hospitals, The Laryngoscope 123 (10) (2013 Oct) 2560–2566. [23] M.I. Orestes, L. Lander, S. Verghese, R.K. Shah, Incidence of laryngospasm and bronchospasm in pediatric adenotonsillectomy, The Laryngoscope 122 (2) (2012 Feb) 425–428. [24] M.E. McCormick, A. Sheyn, M. Haupert, R. Thomas, A.J. Folbe, Predicting complications after adenotonsillectomy in children 3 years old and younger, Int. J. Pediatr. Otorhinolaryngol. 75 (11) (2011 Nov 1) 1391–1394. [25] E.M. Jaryszak, R.K. Shah, C.C. Vanison, L. Lander, S.S. Choi, Polysomnographic variables predictive of adverse respiratory events after pediatric adenotonsillectomy, Arch. Otolaryngol. Head Neck Surg. 137 (1) (2011 Jan 17) 15–18. [26] A. Thongyam, C.L. Marcus, J.L. Lockman, M.A. Cornaglia, A. Caroff, P.R. Gallagher, J. Shults, J.T. Traylor, M.D. Rizzi, L. Elden, Predictors of perioperative complications in higher risk children after adenotonsillectomy for obstructive sleep apnea: a prospective study, Otolaryngology-Head Neck Surg. (Tokyo) 151 (6) (2014 Dec) 1046–1054. [27] A. Shatz, Indications and outcomes of adenoidectomy in infancy, Ann. Otol. Rhinol. Laryngol. 113 (10) (2004 Oct) 835–838. [28] C.M. Baldassari, S. Choi, Assessing adenoid hypertrophy in children: X-ray or nasal endoscopy, The Laryngoscope 124 (7) (2014 Jul 1) 1509–1510. [29] R.G. Elluru, L. Johnson, C.M. Myer III, Electrocautery adenoidectomy compared with curettage and power‐assisted methods, The Laryngoscope 112 (S100) (2002 Aug) 23–25.
[8] P.S. Roland, R.M. Rosenfeld, L.J. Brooks, N.R. Friedman, J. Jones, T.W. Kim, S. Kuhar, R.B. Mitchell, M.D. Seidman, S.H. Sheldon, S. Jones, Clinical practice guideline: polysomnography for sleep-disordered breathing prior to tonsillectomy in children, Otolaryngology-Head Neck Surg. (Tokyo) 145 (1_suppl) (2011 Jul) S1–S5. [9] R.F. Baugh, S.M. Archer, R.B. Mitchell, R.M. Rosenfeld, R. Amin, J.J. Burns, D.H. Darrow, T. Giordano, R.S. Litman, K.K. Li, M.E. Mannix, Clinical practice guideline: tonsillectomy in children, Otolaryngology-Head Neck Surg. 144 (1_suppl) (2011 Jan) S1–S30. [10] R.B. Mitchell, S.M. Archer, S.L. Ishman, R.M. Rosenfeld, S. Coles, S.A. Finestone, N.R. Friedman, T. Giordano, D.M. Hildrew, T.W. Kim, R.M. Lloyd, Clinical practice guideline: tonsillectomy in children (update), Otolaryngology-Head Neck Surg. (Tokyo) 160 (1_suppl) (2019 Feb) S1–S42. [11] C.L. Marcus, L.J. Brooks, S.D. Ward, K.A. Draper, D. Gozal, A.C. Halbower, J. Jones, C. Lehmann, M.S. Schechter, S. Sheldon, R.N. Shiffman, Diagnosis and management of childhood obstructive sleep apnea syndrome, Pediatrics 130 (3) (2012 Sep 1) e714–e755. [12] S.L. Ishman, Evidence-based practice: pediatric obstructive sleep apnea, Otolaryngol. Clin. N. Am. 45 (5) (2012 Oct 1) 1055–1069. [13] O.E. Brown, M.J. Cunningham, A.A. Bull, Tonsillectomy and adenoidectomy inpatient guidelines: recommendations of the AAO-HNS Pediatric Otolaryngology Committee, AAO-HNS Bull. 15 (1996 Sep 13) 1–4. [14] R. Rodman, M. Boehnke, N. Venkatesan, H. Pine, Discharge after tonsillectomy in pediatric sleep apnea patients, Int. J. Pediatr. Otorhinolaryngol. 77 (5) (2013 May 1) 682–685. [15] A.P. Black, S.R. Shott, Is adenoidectomy alone sufficient for the treatment of airway obstruction in children? The Laryngoscope 124 (1) (2014 Jan) 6–7. [16] D.J. Kay, V. Mehta, A.J. Goldsmith, Perioperative adenotonsillectomy management in children: current practices, The Laryngoscope 113 (4) (2003 Apr) 592–597. [17] G.D. Canto, C. Pachêco-Pereira, S. Aydinoz, R. Bhattacharjee, H.L. Tan, L. Kheirandish-Gozal, C. Flores-Mir, D. Gozal, Adenotonsillectomy complications: a meta-analysis, Pediatrics 136 (4) (2015 Oct 1) 702–718. [18] C.L. Marcus, D. Chapman, S.D. Ward, S.A. McColley, C.T. Herrerias, P.C. Stillwell, M. Howenstine, M.J. Light, D.A. Schaeffer, J.S. Wagener, L.N. Laskosz, Clinical practice guideline: diagnosis and management of childhood obstructive sleep apnea
8
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Pediatric adenoidectomy in the very young child and indications for postoperative inpatient admission
T
Stephen R. Chorneya,b, Julia F. Daileya, Karen B. Zura,b,∗ a b
Division of Otolaryngology, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, 19104, USA Department of Otorhinolaryngology - Head and Neck Surgery, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, 19104, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: Adenoidectomy Obstructive sleep apnea Outpatient Postoperative complications Sleep disordered breathing Sleep medicine
Objectives: To determine the rate of significant respiratory events following adenoidectomy in young patients and to identify factors that would prompt inpatient admission postoperatively. Methods: A retrospective chart review was performed of consecutive adenoidectomy surgeries at a high-volume, tertiary-care children's hospital between 2016 and 2018. Children under 3.5 years of age who had surgery for obstructive symptoms were included. Patients were grouped by age (youngest ≤1.5 years, middle 1.6–2.5 years, and oldest 2.6–3.5 years). We excluded patients having revision surgery, a concomitant tonsillectomy, or additional major surgical procedure. Results: There were 353 patients that met inclusion criteria. The three age groups were similar with respect to all characteristics except age (p < .001), body mass index (p < .001), and percentage of Black or African American children (p = .02). Patients under 1.5 years more often had preoperative polysomnography (p = .02) with a lower oxygen saturation nadir (p = .04), and were more likely to have surgery for obstructive sleep apnea (p < .001). No differences were found between age groups with respect to recovery room issues, nurse triage calls, or readmissions within 30 days of surgery. An elective admission rate in the cohort was 35.1%, and this was age-group dependent with 79.5% of the youngest group being admitted (p < .001). On admission, 16.9% of all patients had admission events requiring positive pressure support, intensive care unit admission, or prolonged hospitalization, which was similar across all age groups (p = .67). Events were more common in younger patients (17 mos. vs 20 mos., p = .07), those with more comorbidities (74.8% vs 51.5%, p = .06) and significantly higher in those with severe preoperative polysomnogram variables (p < .001). Based on multivariate regression analysis, younger children (OR: 13.7, 95% CI: 6.5 - 29.0, p < .001) or children with an AHI over 5 events/hr (OR: 32.3, 95% CI: 3.4 - 303.2, p = .005) were more likely to have significant events on admission. Conclusions: Significant respiratory events are uncommon after adenoidectomy for obstructive symptoms, even in very young children. However, for children under 1.5 years of age or those with AHI scores above 5 events/hr, postoperative admission for monitoring is recommended. Clinical judgement should be used when considering outpatient surgery for older children or those with comorbidities.
1. Introduction Adenoidectomy is the third most common ambulatory procedure in pediatric otolaryngology, with an estimated 129,000 procedures annually in the United States [1]. With very low postoperative complication rates following adenoid surgery [2–5] short monitoring and same-day discharge is a reasonable approach for most children [6]. The trend towards more judicious use of inpatient surgical services over the last 10–15 years has resulted in 99.4% of adenoidectomies in 2006 being performed on an outpatient basis [7].
In the pediatric population, sleep disordered breathing is most commonly due to tonsillar and adenoid tissue hypertrophy [8]. While the American Academy of Otolaryngology - Head and Neck Surgery (AAO-HNS) defines tonsillectomy as a surgical procedure that often, but not always, includes the removal of the adenoid with the tonsils, the AAO-HNS guideline on pediatric tonsillectomy does not apply to children younger than age 2 or to children having adenoidectomy alone [9,10]. Although performing an adenoidectomy does not address the oropharyngeal obstruction secondary to tonsillar hyperplasia [11], there has never been any level 1 study looking at the efficacy of
∗ Corresponding author. Division of Otolaryngology, Children's Hospital of Philadelphia, Philadelphia, 3401 Civic Center Blvd., 1 Wood ENT, Pennsylvania, 19104, USA. E-mail addresses: [email protected] (S.R. Chorney), [email protected] (J.F. Dailey), [email protected] (K.B. Zur).
https://doi.org/10.1016/j.ijporl.2019.109796 Received 19 September 2019; Received in revised form 22 November 2019; Accepted 22 November 2019 Available online 25 November 2019 0165-5876/ © 2019 Published by Elsevier B.V.
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
Table 1 Patient characteristics adenoidectomies for obstructive symptoms grouped by age at surgery: 2016–2018.
Total No. (%) Mean age at surgery, yr. (SD) Gender, No. (%) Male Female Gestational age, wk. (SD) NICU admission, No. (%) NICU intubation, No. (%) Body mass index, kg/m2 (SD) Race, No. (%) White Black or African American Asian, Indian, Othera Comorbidities, No. (%) Reactive Airway Disease Otherb None
All Patients
≤1.5 Years
1.6–2.5 Years
2.6–3.5 Years
P Value
353 2.2 (0.7)
73 (20.7) 1.2 (0.3)
164 (46.4) 2.0 (0.3)
116 (32.9) 3.0 (0.3)
< .001
236 (66.9) 117 (33.1) 37.6 (3.5) 89 (25.2) 24 (6.8) 16.8 (1.7)
48 (65.8) 25 (34.2) 37.8 (3.2) 20 (27.4) 4 (5.5) 17.2 (1.7)
109 (66.5) 55 (33.5) 37.7 (3.8) 37 (22.6) 10 (6.1) 17.1 (1.6)
79 (68.1) 37 (31.9) 37.5 (3.5) 32 (27.6) 10 (8.6) 16.3 (1.8)
.95
157 (44.5) 139 (39.4) 57 (16.1)
29 (39.7) 36 (49.3) 8 (11.0)
74 (45.1) 69 (42.1) 21 (12.8)
54 (46.6) 34 (29.3) 28 (24.1)
.02
113 (32.0) 60 (17.0) 180 (51.0)
25 (34.2) 13 (17.8) 35 (47.9)
52 (31.7) 20 (12.2) 92 (56.1)
36 (31.0) 27 (23.3) 53 (45.7)
.15
.84 .57 .74 < .001
SD: standard deviation. P values < 0.05 represent a statistically significant result. a Other race includes small numbers of patients who refused to identify on demographics. b Other comorbidities: Neuromuscular Disorder, Cardiopulmonary Disease, Hematological Disorder, Craniofacial Syndrome, and Trisomy 21.
tube placement, ear exam under anesthesia, cerumen removal, or tympanostomy tube removal with or without paper patch) or minor otolaryngology procedure (nasal cautery, lingual frenotomy, or skin tag excision). Records were analyzed for patient demographics including date of birth, gestational age, gender, body mass index, race, history of NICU admission, and history of NICU intubation. Comorbidities were recorded and grouped according to category. The clinical work-up included whether a lateral X-ray or a nasopharyngoscopy was obtained and these were included along with the reported size of the adenoid. Preoperative polysomnography (PSG), if available, was recorded with Apnea-Hypopnea Index, nadir Oxygen saturation, and peak end tidal CO2 levels. Operative reports were assessed for date of procedure, surgical procedure, surgical indication, intraoperative complications, adenoidectomy technique, and size of adenoid tissue. Notes from physician and nursing staff were included in order to identify any perioperative events or additional support in the post anesthesia care unit (PACU), floor, or intensive care unit. Progress notes and vitals were included to assess for respiratory and non-respiratory events. Interventions, if needed, were also recorded from the electronic progress notes. Location of admission (floor vs. intensive care unit) as well as duration of hospitalization were noted. Upon discharge, if a patient's family contacted our division's nursing answering service, this was documented up to 30 days. The nature of the call was recorded if it was related to the adenoid procedure and not the minor otologic or otolaryngological procedures. Readmissions during the first 30 days were also included, along with the indication, outcomes, and disposition of the patient. Data was analyzed using SPSS® Statistics for Windows (Version 25.0 Armonk, NY: IBM Corp.). Continuous variables were expressed in means with standard deviation. Categorical values were represented by the absolute number along with percentage. Tests for statistical significance were performed by using student's t-testing and analysis of variance (ANOVA) testing where appropriate. Fisher's exact testing was utilized for categorical values. Finally, multinomial logistical regression analyses were performed for various risk factors and the influence they had on significant admission events. Regression outputs were given in odds ratios along with 95% confidence intervals. Significance levels were set at p < 0.05.
adenoidectomy alone for the treatment of obstructive sleep apnea (OSA) or sleep disordered breathing (SDB), which may be a reasonable option for select children [12]. In 1996, the AAO-HNS recommended outpatient tonsillectomy for healthy children with an American Society of Anesthesiologists classification I or II, without evidence of obstructive sleep apnea-hypopnea syndrome (OSAHS), and age older than 3 years [13]. Surgeons must be concerned with patients who are at higher risk for respiratory complications postoperatively. These complications could include desaturations secondary to increased swelling and secretions, apneic events, or pulmonary edema [14]. The question remains whether children should be monitored after adenoidectomy for obstructive symptoms similarly to children who undergo adenotonsillectomy (T&A). Adenoid surgery is shorter in duration, does not involve the oropharynx, and is managed without narcotic pain medication at our institution. The main objective of this study is to characterize complications following adenoidectomy without tonsillectomy for obstructive symptoms in young children. Secondary goals are to identify which children benefit from inpatient admission following surgery in order to monitor for respiratory events. This large retrospective review will also provide meaningful data on the preoperative work-up, the adenoid technique, as well as the frequency of nurse triage calls and readmissions following adenoidectomy within the first 30 days. 2. Methods A retrospective chart review was approved with exemption by the Institutional Review Board at Children's Hospital of Philadelphia. Patients under the age of 3.5 years at the time of surgery were included based on Current Procedural Terminology (CPT) code 42830 (adenoidectomy, primary; under age 12). Only patients with obstructive symptoms were included. Patients who underwent an adenoidectomy for recurrent ear infections, chronic adenoiditis, or otitis without obstructive symptoms were excluded. Consecutive electronic medical records of children who had surgery at our main hospital operating room between January 1, 2016 and July 1, 2018 were reviewed. If a patient had a previous adenoidectomy, a concomitant tonsillectomy, an airway intervention such as bronchoscopy, or an additional major surgical procedure, they were excluded from our analysis. Patients who were tracheostomy or ventilatory dependent were not included. An additional minor procedure with adenoidectomy was acceptable, such as a minor otologic procedure (bilateral myringotomy with tympanostomy 2
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
3. Results
middle age group had a suction cautery adenoidectomy and 52.6% of the oldest group had a microdebrider adenoidectomy (p < .001) (Table 3). There were no recorded intraoperative complications. Events during the perioperative period and the postoperative hospitalization were characterized as minor or significant and were overwhelmingly respiratory-related. Issues in the PACU after adenoidectomy surgery were uncommon, occurring in 8.2% of patients (3.4% significant, 4.8% minor). No differences were seen in these events between age groups (p = .41). There were three episodes of bronchospasm (0.9%), two in the youngest and one in the middle age group, and each was managed with albuterol. There was no laryngospasm and no patient required reintubation. Overall, the planned admission rate was 35.1%, which was age-group dependent as 79.5% of the youngest patients were admitted (p < .001). On admission, 16.9% of patients had a significant event, and this was similar across all age groups (p = .67). Nurse triage calls (27.8%, p = .97), emergency department visits (4.0%), and readmissions within 30 days (0.8%) were not significantly different between groups (p = .51) (Table 4). Not shown was one minor bleed in the PACU that self-resolved and four nurse calls for bleeds (three in the middle group and one in the oldest), with one requiring inpatient readmission for 2 days after bedside packing placement (middle age group). No bleeding occurred in patients with bleeding disorders or with nasal cautery procedures. On readmission, three patients were admitted longer than 24 hours. The most serious complication occurred in a 14 month-old, full term child with reactive airway disease who was readmitted with pneumonia for 8 days requiring video-assisted thoracoscopic surgery (VATS) with chest tube placement for pleural effusion and empyema. Poor oral intake was not an issue for any children in the immediate postoperative period, however we did have four children requiring admission to the ED for hydration and pain control for inadequate oral intake within 30 days. There were no cases of Grisel's syndrome in our cohort. When we looked at characteristics of patients scheduled to be admitted after adenoidectomy, they tended to be younger (1.6 yrs. vs. 2.4 yrs, p < .001), have a higher body mass index (17.2 kg/m2 vs 16.6 kg/m2, p = .01), and were more likely to be Black or African American (49.2% vs 34.1%, p = .001). Additionally, admitted patients with a preoperative sleep study had a higher mean AHI, lower mean nadir O2 saturation, and higher peak end-tidal CO2 (Table 5). Patients with more eventful admissions (i.e. placement of nasopharyngeal airway, need for positive pressure, admission to the ICU, or more prolonged admission) were younger (1.4 yrs vs 1.7 yrs, p = .07), had more comorbidities (74.8% vs 51.5%, p = .06), were more likely to be White (61.9% vs 39.8%, p = .001), and had more severe grouping of AHI, O2 nadir, and end tidal CO2 on PSG (p < .001). Interestingly, while age at surgery was lower, history of NICU admission was higher, and presence of comorbidities were higher in the patients with complicated admissions, these failed to meet statistical significance (Table 6). Subgroup analysis was performed on the youngest group, focusing on events in children with and without a preoperative polysomnogram
There were 353 patients between January 1, 2016 and July 1, 2018 who met inclusion criteria. Table 1 shows the patient characteristics of the study population as well as the characteristics of patients in our three age groups: youngest (≤1.5 years), middle (1.6–2.5 years), and oldest (2.6–3.5 years). The average patient was 2.2 (SD: 0.7) years of age at surgery, with a mean gestational age of 37.6 (SD: 3.5) weeks. Our population included 49.0% of patients diagnosed with at least one comorbidity, with reactive airway disease (32.0%) representing the majority. Not listed, but grouped into other comorbidities, were: Neuromuscular Disorder (4.8%), Cardiopulmonary Disease (4.0%), Hematological Disorder (3.7%), Craniofacial Syndrome (2.0%), and Trisomy 21 (1.7%). Our three population groups were statistically similar with respect to all characteristics except age (p < .001), body mass index (p < .001), and race, with a higher frequency of Black or African American children in the younger group (49.3%, p = .02). The majority of our patients had a preoperative lateral neck X-ray showing moderate or fully obstructing adenoids in 87.7% of children. Nasopharyngoscopy was obtained less frequently for 11.0% of patients, with 89.7% having moderate or fully obstructing adenoids. In our group, 21.5% of children had a preoperative PSG, with a mean AHI of 9.8 (SD: 15.9), mean nadir oxygen saturation of 86.8% (SD: 5.0), and mean peak end tidal CO2 of 52.0 Torr (SD: 5.6) (Table 2). When comparing age groups, we found that nasopharyngoscopy was performed more commonly in the youngest group (p < .001), and the middle group had a higher ratio of non-obstructive and mildly obstructive adenoids on nasopharyngoscopy (p < .001). Additionally, a larger percentage (34.2%) of the youngest age group obtained a preoperative PSG (p = .02) and had a lower mean Oxygen saturation nadir (84.7%, p = .04) than the other two groups. Surgical indications in our study population were adenoid hypertrophy (92.1%), nasal obstruction (46.5%), sleep disordered breathing (46.5%), snoring (24.6%), and obstructive sleep apnea (16.4%) (Table 3). Indications were similar across age groups except for obstructive sleep apnea, which was performed for 32.9% of the youngest group, compared to 11.6% and 12.9% (p < .001). The majority of our patients (66.9%) underwent adenoidectomy with a minor otologic surgery, namely bilateral myringotomy with tympanostomy tube placement. Adenoidectomy alone was performed in 31.2% of cases and adenoidectomy with a minor otolaryngological procedure performed in 2.0% of cases. Older patients were less likely to have a minor otologic procedure and more likely to have an adenoidectomy alone (p = .01). Moderate or fully obstructing adenoids were found intraoperatively in 97.5% of our patients. The three surgical techniques used in our institution were microdebrider (37.4%), suction cautery (36.0%), and coblation (26.6%). The choice of technique is surgeon-dependent, with 14 surgeons in our group during the study period. Significantly more coblation adenoidectomy occurred in the youngest patients (38.4%), while 47.0% of the Table 2 Clinical evaluation adenoidectomies grouped by age at surgery: 2016–2018.
Lateral Neck X-ray, No. (%) Non/Mildly Obstructive Moderate/Fully Obstructive Nasopharyngoscopy, No. (%) Non/Mildly Obstructive Moderate/Fully Obstructive Polysomnography, No. (%) Apnea Hypopnea Index, Mean AHI (SD) Nadir Oxygen Saturation, Mean % (SD) Peak End-Tidal CO2, Mean Torr (SD)
All Patients
≤1.5 Years
1.6–2.5 Years
2.6–3.5 Years
P Value
212 (60.1) 26 (12.3) 186 (87.7) 39 (11.0) 4 (10.3) 35 (89.7) 76 (21.5) 9.8 (15.9) 86.8 (5.0) 52.0 (5.6)
36 (49.3) 4 (11.1) 32 (88.8) 22 (30.1) 2 (9.1) 20 (90.9) 25 (34.2) 12.7 (12.1) 84.7 (5.1) 52.9 (6.6)
102 (62.2) 14 (13.7) 88 (86.3) 8 (4.9) 2 (25.0) 6 (75.0) 29 (17.7) 9.1 (21.9) 87.9 (4.7) 52.0 (5.0)
74 (63.8) 8 (10.8) 66 (89.2) 9 (7.8) 0 (0) 9 (100) 22 (19.0) 7.5 (8.8) 87.9 (4.9) 51.1 (5.2)
.11 .32
SD: standard deviation. P values < 0.05 represent a statistically significant result. 3
< .001 < .001 .02 .52 .04 .58
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
Table 3 Surgical data adenoidectomies grouped by age at surgery: 2016–2018. All Patients
≤1.5 Years
1.6–2.5 Years
2.6–3.5 Years
P Value
325 (92.1) 164 (46.5) 164 (46.5) 87 (24.6) 58 (16.4)
66 25 34 17 24
(90.4) (34.2) (46.6) (23.3) (32.9)
151 (92.1) 82 (50.0) 84 (51.2) 39 (23.8) 19 (11.6)
108 (93.1) 57 (49.1) 46 (39.7) 31 (26.7) 15 (12.9)
.82 .06 .16 .83 < .001
236 (66.9) 110 (31.2) 7 (2.0)
51 (69.9) 20 (27.4) 2 (2.7)
118 (72.0) 46 (28.0) 0 (0)
67 (57.8) 44 (37.9) 5 (4.3)
.01
9 (2.5) 344 (97.5)
2 (2.7) 71 (97.3)
5 (3.0) 159 (97.0)
2 (1.7) 114 (98.3)
.83
132 (37.4) 127 (36.0) 94 (26.6)
26 (35.6) 19 (26.0) 28 (38.4)
45 (27.4) 77 (47.0) 42 (25.6)
61 (52.6) 31 (26.7) 24 (20.7)
< .001
a
Surgical Indication , No. (%) Adenoid Hypertrophy Nasal Obstruction Sleep Disordered Breathing Snoring Obstructive Sleep Apnea Surgical Procedure, No. (%) Adenoidectomy & minor ear surgeryb Adenoidectomy only Adenoidectomy & minor ENT surgeryc Description of Adenoids, No. (%) Non/Mildly Obstructive Moderate/Fully Obstructive Adenoidectomy Technique, No. (%) Microdebrider Suction Cautery Coblation
P values < 0.05 represent a statistically significant result. a Patients may have multiple surgical indications. b Minor otologic surgery: bilateral myringotomy with tympanostomy tube placement, ear exam under anesthesia, cerumen removal, or tympanostomy tube removal with or without paper patch. c Minor ENT surgery: nasal cautery, lingual frenotomy, or skin tag excision. Table 4 Postoperative experience adenoidectomies grouped by age at surgery: 2016–2018.
Issues in PACU, No. (%) Significant respiratory eventa Minor respiratory eventb Admission Rate (planned), No. (%) Significant admission event, No. (%)c Nurse Triage Calls, No. (%) Fever, URI Pain, poor oral intake Bleeding Otherd Readmission Within 30 Days, No. (%) Inpatient readmissione Emergency department observationf
All Patients
≤1.5 Years
1.6–2.5 Years
2.6–3.5 Years
P Value
12 (3.4) 17 (4.8) 124 (35.1) 21 (16.9)
1 (1.4) 6 (8.2) 58 (79.5) 12 (20.7)
6 (3.7) 8 (4.9) 51 (31.1) 7 (13.7)
5 (4.3) 3 (2.6) 15 (12.9) 2 (13.3)
.41
74 (21.0) 18 (5.1) 4 (1.1) 2 (0.6)
16 (21.9) 3 (4.1) 0 1 (1.4)
34 (20.7) 8 (4.9) 3 (1.8) 1 (0.6)
24 (20.7) 7 (6.0) 1 (0.9) 0
.97
3 (0.8) 14 (4.0)
1 (1.4) 2 (2.7)
2 (1.2) 9 (5.5)
0 3 (2.6)
.51
< .001 .67
URI: Upper respiratory tract infection. P values < 0.05 represent a statistically significant result. a Significant airway event: Bronchospasm, oxygen desaturation below 90%, use of high-flow nasal cannula, bag-mask ventilation, use of Continuous Positive Airway Pressure (CPAP), and/or placement of nasopharyngeal airway. b Minor airway event: Oxygen desaturation to 90% with or without placement of low flow oxygen delivery such as nasal cannula, facemask, or blow-by. c Significant admission event: Oxygen desaturation below 85% requiring: use of high flow nasal cannula, bag-mask ventilation, use of Continuous Positive Airway Pressure (CPAP), and/or placement of nasopharyngeal airway; admission to the Intensive Care Unit (ICU) or admission beyond 24 h after surgery. d Other triage calls include: nausea, vomiting, constipation, diarrhea, or neck pain. e Inpatient readmission: significant events that required additional interventions, admission to the floor, or admission beyond 24 h. f Emergency Department observation include patients with pain, nausea, fevers, or poor PO who were monitored for less than 24 h for analgesia and intravenous hydration.
chronic or recurrent otitis media, chronic or recurrent sinusitis, and upper airway obstruction, which includes nasal obstruction, mouth breathing, or snoring [15]. Performing an adenoidectomy alone is a reasonable option for obstructive symptoms such as sleep disordered breathing or obstructive sleep apnea in select children. A survey of AAO-HNS members from New York and the American Society of Pediatric Otolaryngology (ASPO) determined that most surgeons would perform an adenoidectomy alone in a child with OSA and enlarged adenoids, but with small tonsils [16]. Adenoidectomy mitigates early postoperative complications following adenotonsillectomy, including respiratory compromise, hemorrhage, pain, fever, nausea and vomiting, and dehydration [17]. The AAO-HNS clinical practice guideline recommends T&A for OSA [9,10]. Consequently, much of the literature does not separate tonsil and adenoid surgery with respect to complications and the role various risk factors have on respiratory events. We
(Table 7). In children under 1.5 years, age at surgery (p = .81), gender (p = .14), gestational age (p = .65), BMI (p = .57), and presence of comorbidities (p = .41) were similar between those who did and did not have a PSG. While we did admit children more often with a preoperative PSG in this age group (92.0% vs 72.9%, p = .05), rates and severity of events during admission were similar. Finally, multivariate logistical analysis was performed looking at risk factors for children who had significant events on admission (Table 8). Among admitted children, the development of an event was predicted by younger age at surgery (OR: 13.7, 95% CI: 6.5–29.0, p < .001) or an AHI above 5 events/hr (OR: 32.3, 95% CI: 3.4–303.2, p = .005). 4. Discussion Adenoidectomy without tonsillectomy is indicated for children with 4
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
management for patients who are obese, have complicated medical histories, have severe OSA, or are younger than 3 years [9,18]. Patients younger than 3 years and those with multiple comorbidities should be monitored in the hospital because they have relatively high rates of airway complications [19,20]. One group found that children under 3 years of age have a two-fold risk of respiratory complications compared to children 3–6 years [21]. Consequently, admission rates following T& A at pediatric hospitals range from over 94% for children under 2 years with OSA and at least one medical comorbidity, to 14% for children older than 5 years without OSA or comorbidities. However, there still remains significant practice variability even after controlling for age, comorbidities, and OSA [22]. Based on our results, children under 1.5 years of age who have an adenoidectomy alone for obstructive symptoms should be monitored as an inpatient. We did not identify any significant difference between our three age groups with respect to immediate issues in the PACU, frequency of nurse triage calls, or readmissions within 30 days after surgery. Nonrespiratory events such as nausea, vomiting, or poor oral intake were extremely rare. From a bleeding standpoint, there was one self-limited bleed in the PACU immediately after surgery that resolved with topical oxymetazoline. Within the first week, there were four triage calls for bleeding, with one requiring a 48-hour inpatient readmission after placing non-absorbable packing in the emergency department. The rate of bleeding post adenoidectomy has been previously reported in the literature with rates ranging from 0.22 to 1% [2,4,5]. Since our study looks at a subset of patients with obstructive symptoms and does not reflect our entire patient population who undergo adenoidectomy, we cannot extrapolate the global rate of adenoidectomy bleeding at this institution. Overall, our patients had very few significant respiratory-related events. Of the events we recorded, nearly all of them responded to simple interventions. There were three episodes of bronchospasm (0.9%), two in the youngest and one in the middle age group, but these were managed conservatively with albuterol and supplemental oxygen. There was no laryngospasm and no patient required reintubation. The adenotonsillectomy literature has identified rates of 1.6% for laryngospasm and 1.8% for bronchospasm, with bronchospasm occurring after faster surgeries in patients with underlying asthma [23]. Thirty-two percent of our population had a diagnosis of reactive airway disease and this did not impact respiratory event rates during the recovery period. Postoperative respiratory complications can occur in up to 25% of children with OSA after adenotonsillectomy, including oxygen desaturation, atelectasis, pneumonia, pulmonary edema, pleural effusion, pneumothorax, pneumomediastinum, or inspiratory stridor with increased work of breathing [21]. In a large study, children between 1 and 2 years old with a history of nasal obstruction from adenoid hypertrophy, gastroesophageal reflux disease, prematurity, and/or cardiovascular anomalies were at risk for early complications warranting closer observation [24]. Race, gestational age, and BMI were not correlated with complicated events. Further, all of our surgeries were performed for obstructive symptoms among many medically complex patients. While there was a trend for more comorbidities among patients with significant events, this was not statistically significant (p = .06). Therefore, children with comorbidities should raise concern for possible inpatient complications, but our data did not find a clear correlation for this following adenoidectomy. Polysomnogram severity appeared to predict need for support during the inpatient admission following adenoid surgery among the 21.5% of our patients that obtained a preoperative PSG. We found more events in children with an AHI over 5 events per hour, O2 nadir less than 85%, or peak end tidal CO2 over 50 Torr on preoperative PSG (p < .001). Logistical regression analysis identified patients with an AHI over 5 events per hour as having 31 times greater likelihood of a significant respiratory event. While O2 nadir and peak end tidal CO2 did have the same impact on regression modeling, the rates of complicating events in the higher groups suggest that these children warrant close
Table 5 Admission characteristics adenoidectomies grouped by age at surgery: 2016–2018.
Total, No. (%) Mean age at surgery, yr. (SD) Gender, No. (%) Male Female Gestational age, wk. (SD) Body mass index, kg/m2 (SD) Race, No. (%) White Black or African American Asian, Indian, Othera Comorbidities, No. (%) Reactive Airway Disease Otherb None Sleep Study Parameters Apnea Hypopnea Index, Mean AHI (SD) Less than or equal to 5, No. (%)c Greater than 5, No. (%) Nadir Oxygen Saturation, Mean % (SD) Over 85%, No. (%) Less than or equal to 85%, No. (%) Peak End-Tidal CO2, Mean Torr (SD) Less than 50 Torr, No. (%) Greater or equal to 50 Torr, No. (%)
Planned Admission
Outpatient Surgery
P Value
124 (35.1) 1.6 (0.6)
229 (64.9) 2.4 (0.6)
< .001
83 (66.9) 41 (33.1) 37.4 (3.4) 17.2 (1.8)
153 (66.8) 76 (33.2) 37.7 (3.6) 16.6 (1.7)
.54
54 (43.5) 61 (49.2) 9 (7.3)
103 (45.0) 78 (34.1) 48 (21.0)
.001
40 (32.3) 28 (22.6) 56 (45.2)
73 (31.9) 32 (14.0) 124 (54.1)
.09
14.4 (19.4)
3.4 (3.7)
.002
19 (15.3)
28 (12.2)
< .001
25 (20.2) 85.7 (5.3)
4 (1.7) 88.3 (4.3)
.02
27 (21.8) 17 (13.7)
26 (11.4) 6 (2.6)
< .001
53.2 (6.0)
50.5 (4.7)
.04
13 (10.5) 28 (22.6)
14 (6.1) 18 (7.9)
< .001
.44 .01
SD: standard deviation. P values < 0.05 represent a statistically significant result. a Other race includes small numbers of patients who refused to identify on demographics. b Other comorbidities: Neuromuscular Disorder, Cardiopulmonary Disease, Hematological Disorder, Craniofacial Syndrome, and Trisomy 21. c Four children had an Apnea-Hypopnea Index (AHI) less than 1 event/hr (normal): one in the middle group and three in the older age group.
focus on adenoid-only surgery in patients with many of the same risk factors and surgical indications. Adenoidectomy is a milder intervention than adenotonsillectomy, with minimal risks of oropharyngeal swelling, rare bleeding, shorter anesthetic time, and postoperative pain managed without narcotics. Therefore, we suspect that physicians caring for these children could be more aggressive with discharge practices and potentially perform fewer adenoidectomies in an inpatient operating suite. Our institution currently requires children under 2.5 years of age to have their adenoidectomy performed in the main hospital operating room and not at an ambulatory surgical center. The indication for admission post-adenoidectomy is mostly surgeondependent and based on associated comorbidities or severity of sleep apnea. In our large cohort, the post-adenoidectomy elective admission rate of 35.1% was age-group dependent, with 79.5% of the youngest group being admitted (p < .001). Despite balanced event rates across all three groups (20.7%, 13.7%, 13.3%, p = .67), we found that patients with the most significant events were just under 17 months old compared to around 20 months (p = .07). On multivariate regression analysis, younger patients were 13 times more likely to develop a postoperative respiratory event. Age has historically been an important variable in the decision to admit children after adenotonsillectomy. The American Academy of Pediatrics (AAP) and AAO-HNS suggest similar risk factors for postoperative complications and recommend inpatient
5
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
Table 6 Significant admission events adenoidectomies grouped by age at surgery: 2016–2018.
Total, No. (%) Mean age at surgery, yr. (SD) Gender, No. (%) Male Female Gestational age, wk. (SD) Body mass index, kg/m2 (SD) Race, No. (%) White Black or African American Asian, Indian, Othera Comorbidities, No. (%) Reactive Airway Disease Otherb None Sleep Study Parameters Apnea Hypopnea Index, Mean AHI (SD) Less than or equal to 5, No. (%) Greater than 5, No. (%) Nadir Oxygen Saturation, Mean % (SD) Over 85%, No. (%) Less than or equal to 85%, No. (%) Peak End-Tidal CO2, Mean Torr (SD) Less than 50 Torr, No. (%) Greater or equal to 50 Torr, No. (%)
Admissions Without Significant Events
Admissions With Significant Events
P Value
103 (83.1) 1.7 (0.6)
21 (16.9) 1.4 (0.5)
.07
70 (68.0) 33 (32.0) 37.5 (3.4) 17.2 (1.9)
13 (61.9) 8 (38.1) 37.1 (3.4) 17.1 (1.7)
.83
41 (39.8) 54 (52.4) 41 (39.8)
13 (61.9) 7 (33.3) 1 (4.8)
.001
33 (32.0) 20 (19.4) 50 (48.5)
7 (33.3) 8 (38.1) 6 (28.6)
.06
12.2 (16.3) 15 (14.6) 16 (15.5) 86.4 (4.4) 21 (20.4) 10 (9.7) 53.1 (5.8) 10 (9.7) 20 (19.4)
19.8 (25.4) 4 (19.0) 9 (42.9) 84.2 (6.9) 6 (28.6) 7 (33.3) 53.6 (6.7) 3 (14.3) 8 (38.1)
.24 < .001
.74 .81
.20 < .001 .79 < .001
SD: standard deviation. P values < 0.05 represent a statistically significant result. a Other race includes small numbers of patients who refused to identify on demographics. b Other comorbidities: Neuromuscular Disorder, Cardiopulmonary Disease, Hematological Disorder, Craniofacial Syndrome, and Trisomy 21.
admission was to the ICU, with 80% of these in the under 1.5-year-old age group. Of note, 7 of 10 in the youngest group were discharged from the ICU on postoperative day (POD) 1, while the other 3 were discharged in 48–72 hours. This calls into question the role of ICU admission afterwards in adenoidectomy patients. A previous study has suggested that in children under the age of 12 months who have an adenoidectomy for OSA, postoperative monitoring in the intensive care unit for 24 hours is highly recommended [27]. Depending on staffing and institutional comfort, many of these children could be managed on the floor with close nursing care as most of our ICU patients did not
attention as well. To date, the relationship between PSG variables and adenoidectomy complications have not been well-studied. Looking at adenotonsillectomy outcomes, one group noted that patients with desaturations following T&A requiring supplemental oxygen had significantly higher AHI, (31.8 vs 14.1; p = .001), higher hypopnea index (22.6 vs 8.9; p = .004), and lower nadir oxygen saturation (72% vs 84%; p = .001) [25]. Further, O2 nadir less than 80% or peak CO2 greater than 60 mm Hg on PSG are strong and consistent predictors of postoperative respiratory complications following adenotonsillectomy in children with OSA [26]. For 12 of our patients, postoperative
Table 7 Children under 1.5 years comparing PSG without PSG adenoidectomies: 2016–2018.
Total, No. (%) Mean age at surgery, yr. (SD) Male gender, No. (%) Gestational age, wk. (SD) Body mass index, kg/m2 (SD) Comorbidities, No. (%) Admitted, No. (%) PACU Events, No. (%) PACU Event Details, (No., %)
Admission Events, No. (%) Admission Event Details, (No., %)
No Sleep Study
Preoperative Sleep Study
48 (65.8) 1.2 (0.3) 29 (60.4) 37.9 (3.1) 17.1 (1.7) 24 (50.0%) 35 (72.9) 5 (10.4) · Broncospasm, albuterol given, admitted to floor (1, 2.1%) · O2 NC, weaned on floor (1, 2.1%) · Desats requiring NC or HFNC, with ICU admission (3, 6.3%) 5 (14.2) · ICU, desats to low 70's needing 1L NC, weaned (2, 5.7%) · ICU, NP airway and HFNC for desats to low 80's (1, 2.9%) · Floor, 2L NC weaned for desats to upper 70's-low 80's (2, 5.7%) · Readmit for 8 days for pneumonia, VATS for empyema (1, 2.1%)
25 (34.2) 1.2 (0.3) 19 (76.0) 37.9 (3.4) 17.4 (1.6) 14 (56.0%) 23 (92.0) 2 (8.0) · Bronchospasm in OR, albuterol given, observed in ICU without O2 (1, 4.0%) · Brief desats, 1L NC, weaned on floor (1, 4.0%)
7 (30.4) · ICU, 2L NC weaned to RA (1, 4.3%) · ICU, desats to 82% repositioning, no O2 needed (1, 4.3%) · ICU,no desats or O2 (5, 21.7%)
P value
.81 .14 .65 .57 .41 .05 1.00
.13
ICU = intensive care unit, NC = nasal cannula, NP = nasopharyngeal, HFNC = high flow nasal cannula, desats = desaturations, VATS= Video-Assisted Thoracoscopic Surgery, RA = room air. 6
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
missed in other children who met ambulatory surgery center criteria. Third, only children with obstructive symptoms were included, which was strategic from a study design as we made the assumption that children with obstructive symptoms may have more difficulty with recovery compared to the patients with no pre-existing sleep issues. However, this could lead to an inability to generalize to all children who have an adenoidectomy. To analyze whether we can liberalize the recommendation for admission in the middle or youngest age groups, we will be studying patients who had an adenoidectomy for infection/ adenoiditis. Fourth, grouping various comorbidities due to small numbers limits generalizability and analysis, which may be an important area of post-operative complications. Finally, the ability to determine respiratory complications and measurements of events are limited by the accuracy of the documentation of clinical staff. As such, there could be events that are missed if they were not recorded in the patient's electronic record.
Table 8 Multivariate analysis of risks for significant admission events adenoidectomies: 2016–2018.
Younger age at surgery Gender Younger gestational age Higher body mass index Race Presence of comorbidities Higher Apnea Hypopnea Index Lower Nadir Oxygen Saturation Higher Peak End-Tidal CO2
Odds Ratio
95% Confidence Interval
P Value
13.7 1.1 1.0 0.9 3.5 1.5 32.3 0.6 1.8
(6.5–29.0) (0.5–2.4) (0.9–1.2) (0.7–1.1) (0.7–17.1) (0.6–3.7) (3.4–303.2) (0.1–3.9) (0.3–10.0)
< .001 .39 .81 .40 .09 .25 .005 .68 .46
P values < 0.05 represent a statistically significant result.
require intensive services aside from two patients requiring CPAP overnight. The vast majority of our children had a lateral neck X-ray (60.1%) but a flexible nasopharyngoscopy was performed for 11% of our patients. Notably, our youngest age group was more likely to have a nasopharyngoscopy (30.1%, p < .001). The reasons for this are unclear, but may be related to practitioner preference, whether X-ray studies were previously done by referring clinicians, ease of examination, parental choice (both financial and related to their concern about performing a nasopharyngoscopy on their child), and less radiation. There are inherent issues with measuring size of adenoids, but in our institution, we rely heavily on the lateral neck film. A recent best practice review stated that flexible nasal endoscopy is well tolerated in most children, allows for direct visualization of the adenoid, and is comparable to lateral neck X-ray in cost. Nasal endoscopy also avoids variation due to poor patient position and radiation exposure of lateral neck films. The authors recommend a flexible nasal endoscopy as the best initial choice for evaluation of adenoid size when the tonsils are small, with a lateral neck X-ray reserved for children who are unable to cooperate with a flexible nasal endoscopic exam [28]. Regardless of assessment technique, the decision to perform adenoidectomy alone in our population was nearly universally a result of small tonsils on exam with evidence of large adenoids from X-ray or nasal endoscopy. Microdebrider and suction cautery were the two most common surgical techniques, with a slightly smaller amount of coblation adenoidectomy. We did find some age-group differences, with 38.4% of the youngest patients having coblation adenoidectomy, 47.0% of the middle age group having suction cautery adenoidectomy, and 52.6% of the oldest group having a microdebrider adenoidectomy (p < .001). This difference likely represents practitioner preferences as opposed to a meaningful relationship to the age of the patient. Nonetheless, each technique is used routinely in all age groups. A review of adenoidectomy techniques found that electrocautery adenoidectomy has an advantage with regards to minimizing operative time and blood loss at a lower cost than power-assisted adenoidectomy [29]. The merits and drawbacks to each approach are beyond the scope of this paper, but we did not identify any intraoperative or postoperative complications and therefore find a role for all three techniques. There are several limitations to the results of this study. First, the retrospective nature of the data introduces selection bias particularly with respect to younger children having more documented OSA on PSG. Reluctance to perform tonsillectomy in young patients for risk of severe postoperative complications may have been a reason for clinicians ordering preoperative PSG more frequently. As a result, older children may have similar polysomnogram severity, but do not appear this way given a lower frequency of PSG ordering. Second, we only looked at patients who had surgery at the main operating room and not in our outpatient surgery centers. The tendency to electively admit younger patients after adenoid surgery in our institution may have overestimated the rate of complications in this group, which we may have
5. Conclusions Adenoidectomy without tonsillectomy has a role in the management of select children with upper airway obstruction. Although postoperative respiratory events are rare, we identified age and severity of PSG variables as predictive of events that would warrant admission. In particular, for children under 1.5 years or those with severe PSG variables (AHI above 5 events/hr, nadir O2 below 85%, or peak end tidal CO2 above 55 Torr), inpatient monitoring for respiratory complications is recommended. However, since only a fraction of the patients in our cohort had a preoperative sleep study, we recommend that clinical judgement should be used when considering outpatient surgery in the youngest age group. Furthermore, for patients who are older than 1.5 years of age, consideration can be made for performing an adenoidectomy at a surgical center setting barring additional comorbidities that would exclude their eligibility. Our results provide one of the largest single-center cohorts published on events following adenoidectomy for obstructive symptoms. Future studies should continue to define patient characteristics that determine appropriate time for discharge after adenoidectomy in young children. Declaration of competing interest The authors have no conflicts of interest to report. Acknowledgements The authors would like to acknowledge Terri Giordano DNP, CRNP, CORLN who provided invaluable support during the design and Institutional Review Board submission of this research. References [1] N. Bhattacharyya, Ambulatory pediatric otolaryngologic procedures in the United States: characteristics and perioperative safety, The Laryngoscope 120 (4) (2010 Apr) 821–825. [2] J.P. Windfuhr, Y.S. Chen, Post-tonsillectomy and-adenoidectomy hemorrhage in nonselected patients, Ann. Otol. Rhinol. Laryngol. 112 (1) (2003 Jan) 63–70. [3] A. Hunt, M. Karela, P.J. Robb, Day-case adenoidectomy: outcomes are improved using suction coagulation and prophylactic anti-emetic treatment, Int. J. Pediatr. Otorhinolaryngol. 69 (12) (2005 Dec 1) 1629–1633. [4] J.P. Windfuhr, Y.S. Chen, S. Remmert, Hemorrhage following tonsillectomy and adenoidectomy in 15,218 patients, Otolaryngology-Head Neck Surg. (Tokyo) 132 (2) (2005 Feb) 281–286. [5] K. Thomas, D. Boeger, J. Buentzel, D. Esser, K. Hoffmann, P. Jecker, A. Mueller, G. Radtke, K. Geißler, M. Finkensieper, O. Guntinas-Lichius, Pediatric adenoidectomy: a population-based regional study on epidemiology and outcome, Int. J. Pediatr. Otorhinolaryngol. 77 (10) (2013 Oct 1) 1716–1720. [6] S. Habib, K. Solan, P. Das, Reducing observation time in children post‐adenoidectomy, Clin. Otolaryngol. 44 (2) (2019 Mar) 204. [7] N. Bhattacharyya, H.W. Lin, Changes and consistencies in the epidemiology of pediatric adenotonsillar surgery, 1996–2006, Otolaryngology-Head Neck Surg. (Tokyo) 143 (5) (2010 Nov) 680–684.
7
International Journal of Pediatric Otorhinolaryngology 130 (2020) 109796
S.R. Chorney, et al.
syndrome, Pediatrics 109 (4) (2002) 704–712. [19] M. Amoils, K.W. Chang, O. Saynina, P.H. Wise, A. Honkanen, Postoperative complications in pediatric tonsillectomy and adenoidectomy in ambulatory vs inpatient settings, JAMA Otolaryngology-Head Neck Surg. 142 (4) (2016 Apr 1) 344–350. [20] C.M. Lawlor, C.A. Riley, J.M. Carter, K.H. Rodriguez, Association between age and weight as risk factors for complication after tonsillectomy in healthy children, JAMA Otolaryngology-Head Neck Surg. 144 (5) (2018 May 1) 399–405. [21] M.M. Statham, R.G. Elluru, R. Buncher, M. Kalra, Adenotonsillectomy for obstructive sleep apnea syndrome in young children: prevalence of pulmonary complications, Arch. Otolaryngol. Head Neck Surg. 132 (5) (2006 May 1) 476–480. [22] S.S. Goyal, R. Shah, D.W. Roberson, M.L. Schwartz, Variation in post‐adenotonsillectomy admission practices in 24 pediatric hospitals, The Laryngoscope 123 (10) (2013 Oct) 2560–2566. [23] M.I. Orestes, L. Lander, S. Verghese, R.K. Shah, Incidence of laryngospasm and bronchospasm in pediatric adenotonsillectomy, The Laryngoscope 122 (2) (2012 Feb) 425–428. [24] M.E. McCormick, A. Sheyn, M. Haupert, R. Thomas, A.J. Folbe, Predicting complications after adenotonsillectomy in children 3 years old and younger, Int. J. Pediatr. Otorhinolaryngol. 75 (11) (2011 Nov 1) 1391–1394. [25] E.M. Jaryszak, R.K. Shah, C.C. Vanison, L. Lander, S.S. Choi, Polysomnographic variables predictive of adverse respiratory events after pediatric adenotonsillectomy, Arch. Otolaryngol. Head Neck Surg. 137 (1) (2011 Jan 17) 15–18. [26] A. Thongyam, C.L. Marcus, J.L. Lockman, M.A. Cornaglia, A. Caroff, P.R. Gallagher, J. Shults, J.T. Traylor, M.D. Rizzi, L. Elden, Predictors of perioperative complications in higher risk children after adenotonsillectomy for obstructive sleep apnea: a prospective study, Otolaryngology-Head Neck Surg. (Tokyo) 151 (6) (2014 Dec) 1046–1054. [27] A. Shatz, Indications and outcomes of adenoidectomy in infancy, Ann. Otol. Rhinol. Laryngol. 113 (10) (2004 Oct) 835–838. [28] C.M. Baldassari, S. Choi, Assessing adenoid hypertrophy in children: X-ray or nasal endoscopy, The Laryngoscope 124 (7) (2014 Jul 1) 1509–1510. [29] R.G. Elluru, L. Johnson, C.M. Myer III, Electrocautery adenoidectomy compared with curettage and power‐assisted methods, The Laryngoscope 112 (S100) (2002 Aug) 23–25.
[8] P.S. Roland, R.M. Rosenfeld, L.J. Brooks, N.R. Friedman, J. Jones, T.W. Kim, S. Kuhar, R.B. Mitchell, M.D. Seidman, S.H. Sheldon, S. Jones, Clinical practice guideline: polysomnography for sleep-disordered breathing prior to tonsillectomy in children, Otolaryngology-Head Neck Surg. (Tokyo) 145 (1_suppl) (2011 Jul) S1–S5. [9] R.F. Baugh, S.M. Archer, R.B. Mitchell, R.M. Rosenfeld, R. Amin, J.J. Burns, D.H. Darrow, T. Giordano, R.S. Litman, K.K. Li, M.E. Mannix, Clinical practice guideline: tonsillectomy in children, Otolaryngology-Head Neck Surg. 144 (1_suppl) (2011 Jan) S1–S30. [10] R.B. Mitchell, S.M. Archer, S.L. Ishman, R.M. Rosenfeld, S. Coles, S.A. Finestone, N.R. Friedman, T. Giordano, D.M. Hildrew, T.W. Kim, R.M. Lloyd, Clinical practice guideline: tonsillectomy in children (update), Otolaryngology-Head Neck Surg. (Tokyo) 160 (1_suppl) (2019 Feb) S1–S42. [11] C.L. Marcus, L.J. Brooks, S.D. Ward, K.A. Draper, D. Gozal, A.C. Halbower, J. Jones, C. Lehmann, M.S. Schechter, S. Sheldon, R.N. Shiffman, Diagnosis and management of childhood obstructive sleep apnea syndrome, Pediatrics 130 (3) (2012 Sep 1) e714–e755. [12] S.L. Ishman, Evidence-based practice: pediatric obstructive sleep apnea, Otolaryngol. Clin. N. Am. 45 (5) (2012 Oct 1) 1055–1069. [13] O.E. Brown, M.J. Cunningham, A.A. Bull, Tonsillectomy and adenoidectomy inpatient guidelines: recommendations of the AAO-HNS Pediatric Otolaryngology Committee, AAO-HNS Bull. 15 (1996 Sep 13) 1–4. [14] R. Rodman, M. Boehnke, N. Venkatesan, H. Pine, Discharge after tonsillectomy in pediatric sleep apnea patients, Int. J. Pediatr. Otorhinolaryngol. 77 (5) (2013 May 1) 682–685. [15] A.P. Black, S.R. Shott, Is adenoidectomy alone sufficient for the treatment of airway obstruction in children? The Laryngoscope 124 (1) (2014 Jan) 6–7. [16] D.J. Kay, V. Mehta, A.J. Goldsmith, Perioperative adenotonsillectomy management in children: current practices, The Laryngoscope 113 (4) (2003 Apr) 592–597. [17] G.D. Canto, C. Pachêco-Pereira, S. Aydinoz, R. Bhattacharjee, H.L. Tan, L. Kheirandish-Gozal, C. Flores-Mir, D. Gozal, Adenotonsillectomy complications: a meta-analysis, Pediatrics 136 (4) (2015 Oct 1) 702–718. [18] C.L. Marcus, D. Chapman, S.D. Ward, S.A. McColley, C.T. Herrerias, P.C. Stillwell, M. Howenstine, M.J. Light, D.A. Schaeffer, J.S. Wagener, L.N. Laskosz, Clinical practice guideline: diagnosis and management of childhood obstructive sleep apnea
8