- Email: [email protected]
Respiratory events after adenotonsillectomy requiring escalated admission status in children with obstructive sleep apnea
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Respiratory events after adenotonsillectomy requiring escalated admission status in children with obstructive sleep apnea
T
Alexandra M. Arambulaa,b, Deborah X. Xiea,b, Amy S. Whighama,b,c,∗ a
Vanderbilt University School of Medicine, Nashville, TN USA Surgical Outcomes Center for Kids, Vanderbilt University Medical Center, Nashville, TN, USA c Department of Otolaryngology - Head and Neck Surgery, Pediatric Otolaryngology, Vanderbilt University Medical Center, Nashville, TN USA b
A R T I C L E I N F O
A B S T R A C T
Keywords: Adenotonsillectomy Tonsillectomy Obstructive sleep apnea Postoperative complications Intensive care Pediatric
Objectives: To characterize postoperative respiratory complications following adenotonsillectomy (AT) in children with obstructive sleep apnea (OSA) and to identify variables associated with pediatric intensive care unit (PICU) admission. Methods: Retrospective analysis of 133 pediatric OSA patients with prior AT. Assessment of the postoperative hospital course informed patient stratification based on respiratory event severity, PICU admission status, and unscheduled escalation of care. Results: Thirty-six (26.8%) patients were admitted to the PICU. Compared to non-PICU admissions, these patients were significantly younger and with greater preoperative apnea-hypopnea indices, comorbidities, and percentage of post-anesthesia care unit (PACU) time requiring supplemental oxygen. Seventy-one respiratory events occurred in 59 patients, with 60.6% affecting PICU patients. Fifteen severe events occurred, affecting 31% of PICU patients. Of 14 unscheduled escalations of care, 7 were PICU admissions who, compared to planned PICU admissions, spent significantly more time in the PACU and exhibited a trend towards greater PACU time on supplemental oxygen. Conclusions: Pediatric patients requiring post-AT PICU care have more risk factors for respiratory compromise. Total PACU time and total PACU time requiring supplemental oxygen may indicate patient risk for postoperative respiratory complications and need for intensive care. Future work includes prospective determination of appropriate post-AT PICU admission.
1. Introduction Obstructive sleep apnea (OSA) is an increasingly common sleeprelated breathing disorder that affects 1–5% of children [1]. OSA is characterized by disruptions in sleep patterns secondary to partial or complete collapse of the upper airway. Well-known risk factors for OSA include male gender, black race, obesity, craniofacial disorders, neuromuscular disease, and Down syndrome [1–4]. As adenotonsillar hypertrophy (ATH) contributes to a large percentage of pediatric OSA cases, the American Academy of Pediatrics and American Academy of Otolaryngology – Head and Neck Surgery recommend adenotonsillectomy (AT) as the primary treatment modality for children with OSA and ATH, as this procedure leads to significantly improved polysomnographic findings in uncomplicated patients [1,5–7]. One serious complication associated with AT is postoperative respiratory compromise secondary to upper airway obstruction. This risk increases from 0-1.3% to 16–27% with a diagnosis of OSA [8]. The
∗
likelihood of respiratory compromise further increases in high-risk patients, including children with age < 3 years, weight < 5th percentile for age, craniofacial abnormalities, severe OSA as measured on polysomnography (PSG), cardiac disease, prematurity, obesity, and hypotonia [8–10]. Identification of these characteristics in the early 1990s led many hospitals to electively admit high-risk patients to the pediatric intensive care unit (PICU) for airway monitoring after AT. Retrospective reviews indicate that routine postoperative ICU admission may be unnecessary for many of these patients, although no consensus exists [8,11–15]. Our own institution lacks clear guidelines regarding postoperative elective ICU admission. Given the increased financial cost of ICU admission and limited availability of PICU beds, identification of patient variables associated with a need for intensive care is essential in developing a best-practice algorithm. This study investigates post-AT respiratory complications in pediatric OSA patients in order to identify possible factors that may correlate with necessity for an ICU level of care.
Corresponding author. Doctors' Office Tower, 7th Floor, 2200 Children's Way, Nashville, TN 37232-8605, USA. E-mail address: [email protected] (A.S. Whigham).
https://doi.org/10.1016/j.ijporl.2018.01.009 Received 27 September 2017; Received in revised form 10 January 2018; Accepted 14 January 2018 Available online 31 January 2018 0165-5876/ © 2018 Elsevier B.V. All rights reserved.
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
2. Methods
3. Results
This retrospective review was approved by the Vanderbilt University Institutional Review Board (IRB #170372). Patients with diagnosed sleep disordered breathing (SDB) or OSA (clinically or by PSG) managed by the pediatric otolaryngology service at our institution were identified from a clinic database. Inclusion criteria were patients less than 18 years of age, a diagnosis of sleep disordered breathing (SDB) or OSA (clinically or by PSG), evaluation by otolaryngologists at our institution between 2014 and 2015, and prior AT at our institution. Patient demographic information was recorded. Comorbid risk factors were defined as Down syndrome, cerebral palsy, hypotonia, craniofacial abnormalities, neuromuscular disease, or laryngotracheomalacia. Data about each patient's postoperative hospital course were also collected. Though surgeon and anesthesiologist dependent, this institution's postoperative ICU admission criteria include factors identified by Hill et al.: obstructive apnea-hypopnea index (AHI) > 24; age < 24 months; oxygen saturation < 90% on room air in the post-anesthesia care unit (PACU); and intraoperative complications including bronchospasm or laryngospasm requiring treatment [16]. Additional admission criteria include airway obstruction requiring nasal trumpet placement; oxygen requirement exceeding 40% fraction of inspired oxygen (FiO2) in the PACU; craniofacial abnormalities including Down syndrome and Pierre Robin sequence; and neuromuscular/neurodevelopmental disorders such as cerebral palsy and other conditions where patients require frequent suctioning to assist in management of secretions. Respiratory events occurring during the postoperative hospital course, defined as the time period from surgical extubation to discharge, were recorded and stratified into mild and severe events (Table 1). A desaturation requiring multiple interventions was counted as a single respiratory event. Desaturations requiring interventions under both the mild and severe event categories were classified as severe respiratory events. Respiratory events were further categorized based on whether the complications required an escalation of care. This was defined as any unplanned admission to either the hospital floor or PICU. Patients experiencing multiple respiratory events were categorized based on the most severe event. Descriptive statistics were performed and computed as means with standard deviation (SD) for quantitative variables and counts and percentages for categorical variables. Differences between patients who did or did not require PICU admission and between patients who did or did not have an unexpected escalation of care were determined. Unpaired, two-sample t-tests and univariate analysis were used to compare continuous variables, and chi-squared tests were used to compare categorical variables. Statistical significance was set a priori at p < 0.05.
Of 312 pediatric patients with SDB/OSA, 133 met inclusion criteria. Demographic information is shown in Table 2. A small majority of patients were male (56.4%), with this predominance increasing further among PICU patients (69.4%). Over 90% of all patients identified as either white or African American. Preoperative PSG was performed in 105 (78.9%) patients, with obstructive AHIs ranging from 0.6 to 63.5 events/hr. Fifty-seven percent of all patients had a comorbid diagnosis (47% of standard admission [non-PICU] patients, 83% of PICU patients). Following their operation, two patients were directly admitted to the PICU. The remaining 131 patients went to the PACU and were either discharged home (20 patients, 15.3%) or admitted to the hospital floor (77 patients, 58.8%) or the PICU (34 patients, 26.0%). Thus, a total of 36 patients (27.1%) were admitted to the PICU. As seen in Table 2, these patients were significantly younger than non-PICU patients, with higher preoperative AHIs, and with more comorbid risk factors for respiratory compromise. Table 3 contains information on the patients' immediate postoperative and hospital course. Within this group, 7 patients (1 PICU, 6 non-PICU) had missing PACU records. The two groups spent a similar amount of time in the PACU; however, patients admitted to the PICU ultimately spent a significantly longer percentage of this time on supplemental oxygen. They also had significantly longer hospital admissions (p < 0.001), more hospital days on supplemental oxygen (p < 0.001), and more hospital days requiring additional respiratory support (p = 0.041). Seventy-one respiratory events occurred in 59 patients (44.4%), requiring 125 interventions (Table 3). The most common respiratory event was a desaturation (n = 55, 77.5% of all events). The average oxygen saturation for children placed on supplemental oxygen was 82.9 ± 9.8%, with a trend towards lower saturations in the PICU group (Table 3). The lowest recorded saturations were 53% and 50% in the PACU and PICU, respectively. Nearly all desaturations were treated with administration of supplemental oxygen (n = 54, 43.2% of all interventions, Fig. 1). Of all events requiring supplemental oxygen, 61.1% occurred in the PACU (Fig. 2). Additional respiratory interventions were more commonly needed by patients ultimately admitted to the PICU (63 vs. 8 in non-PICU patients). These interventions more often occurred in the PICU compared to the PACU, including placement of a nasal trumpet (n = 13, 46.2% vs. 30.8%), airway suctioning (n = 19, 47.4% vs. 31.6%), and steroid administration (n = 11, 45.5% vs. 27.3%). Fifteen respiratory events (21.1%) were classified as severe, occurring in 15 patients (11 PICU, 4 non-PICU). The remaining 56 mild respiratory events (78.9%) occurred in 50 patients, including 6 of the
Table 1 Stratification of respiratory events. Mild respiratory events
Apnea Stridor Laryngospasm Bronchospasm Oxygen desaturation requiring at least one of the following: Re-initiation of supplemental oxygen Jaw thrust Nasal trumpet placement Airway suctioning
Severe respiratory events
Respiratory distress requiring initiation of rapid response team Oxygen desaturation requiring at least one of the following: Bronchodilator Muscle relaxant Initiation of continuous positive airway pressure (CPAP) Initiation of bilevel positive airway pressure (BiPAP) Intubation
32
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
Table 2 Demographic and clinical patient data.
Age at surgery (years) Sex Female Male Raceb White African-American Asian Hispanic Native Hawaiian/Other Pacific Islander Weight percentile-for-age (%) Pre-AT AHI (events/hr) Number of comorbidities (per patient)
All Patients (n = 133), n (%); mean ± SDa
Non-PICU (n = 97), n (%); mean ± SDa
PICU (n = 36), n (%); mean ± SDa
p value
5.3 ± 4.0
6.1 ± 4.2
3.4 ± 2.5
< 0.001 0.064
58 (43.6%) 75 (56.4%)
47 (48.5%) 50 (51.5%)
11 (30.6%) 25 (69.4%)
90 (67.7%) 32 (24.1%) 3 (2.3%) 6 (4.5%) 1 (0.7%)
69 (71.1%) 19 (19.6%) 3 (3.1%) 5 (5.2%) 1 (1.0%)
21 (58.3%) 13 (36.1%) – 1 (2.8%) –
51.9 ± 36.6 9.4 ± 11.3 1.1 ± 1.2
54.9 ± 36.6 6.0 ± 4.6 0.9 ± 1.1
43.8 ± 35.9 17.3 ± 16.8 1.7 ± 1.2
< 0.001 0.161 0.048 0.557
0.120 < 0.001 < 0.001
Data for all patients, further subdivided by non-PICU or PICU admission status. Non-PICU patients were children requiring only standard admission. PICU patients were children requiring admission to the PICU. P values reflect comparisons between the PICU and non-PICU groups. a Categorical variables are described as n (%); continuous variables are described as mean ± SD. b 1 PICU patient declined to indicate race.
floor rather than same-day discharge, 1 to the PICU rather than sameday discharge, and 2 to the PICU rather than scheduled floor admission. Five of the 15 patients with severe respiratory events (33.3%) required an escalation of care: 1 floor admission rather than same-day discharge and 4 PICU admissions rather than hospital floor admissions (Fig. 3). In total, 14 patients had an unscheduled escalation of care (Table 4). Seven required unplanned PICU admission (19.4% of PICU patients). Seven had an unplanned hospital admission (7.2% of non-PICU patients), but only 2 were admitted for respiratory concerns, accounting for one severe and one mild respiratory event. Other reasons for admission included poor oral intake, nausea and vomiting. Four of the
patients who experienced a severe event. Compared to those experiencing mild respiratory events, patients experiencing severe events had a lower weight percentile-for-age (29.7% vs. 49.1%, p = 0.057). There was a statistically significant difference in length of admission (4.5 vs. 2.1 days, p < 0.001), percent of hospital days requiring supplemental oxygen (60.7% vs. 23.6%, p < 0.001), and percent of total PACU time requiring supplemental oxygen (84% vs. 58%, p = 0.026) between these groups. There were no significant differences based on age, preoperative PSG, comorbid risk factors per patient, or total time spent in the PACU. Of the patients who experienced one or more mild respiratory events, 4 (8.0%) needed an escalation of care: 1 to the hospital
Table 3 Postoperative respiratory events/interventions and oxygen requirements.
Total time in PACU (min) Lowest O2 saturation prior to supplemental O2 administration (%) Total PACU time on O2 (min) % total PACU time on O2 (%) Number of respiratory events Apnea Stridor Oxygen desaturation Laryngospasm Bronchospasm Number of interventionsb Nasal trumpet Jaw thrust Suction Postoperative oxygen Muscle relaxants Bronchodilators Steroids CPAP BiPAP Rapid response Intubation Length of hospital admission (days) [range; median] Proportion of admission days requiring supplemental O2 [range; median] Proportion of admission days requiring other respiratory interventions [range; median]
Non-PICU (n = 97); n (%); mean ± SDa
PICU (n = 36); n (%); mean ± SDa
p value
144.5 ± 119.9 84.6 ± 8.4 43.0 ± 57.5 30.1 ± 29.3 28 (39.4%) – 2 (7.1%) 26 (36.6%) – – 34 (27.2%) 1 (2.9%) – 2 (5.9%) 26 (76.5%) – 3 (8.8%) 1 (2.9%) 1 (2.9%) – – – 1.3 ± 1.4 [0–8; 1] 0.09 ± 0.25 [0–3; 0] 0.04 ± 0.35 [0–4; 0]
158.1 ± 96.2 81.2 ± 10.9 111.4 ± 101.5 70.0 ± 37.9 43 (60.6%) 1 (1.4%) 11 (15.5%) 29 (40.8%) 2 (2.8%) – 91 (73.6%) 12 (13.2%) 5 (5.5%) 17 (18.7%) 28 (30.8%) 2 (2.2%) 5 (5.5%) 10 (11.0%) 5 (5.5%) 2 (2.2%) 2 (2.2%) 3 (3.3%) 3.2 ± 2.7 [1–11; 2] 0.39 ± 0.32 [0–9; 0.625] 0.18 ± 0.34 [0–5; 1]
0.542 0.058 < 0.001 < 0.001
< 0.001 < 0.001 0.041
Clinical data from the immediate postoperative course and entire hospital admission. All percentages are calculated based on total patient population, unless otherwise specified. P values reflect comparisons between the PICU and non-PICU groups. O2 = supplemental oxygen. a Categorical variables are described as n (%); continuous variables are described as mean ± SD. b Percentages for intervention subcategories were calculated from the number of interventions in respective patient groups (non-PICU or PICU).
33
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
statistically significant difference in demographic or clinical characteristics. There was a trend towards greater pre-operative AHIs for patients with scheduled PICU admission (p = 0.056). The postoperative course differed significantly for days on supplemental oxygen above baseline requirements (3.5 vs. 1.2 days, p = 0.009), days requiring additional respiratory interventions (1.6 vs. 1.1 days, p = 0.025), and total time spent in the PACU (225 vs. 144 min, p = 0.047). There was a trend towards greater total PACU time spent on supplemental oxygen (176 vs. 98 min, p = 0.071) in the group with unscheduled admissions, but there was no difference in number of severe events per patient (0.24 [median 0] vs. 0.57 [median 1] events, p = 0.294). 4. Discussion
Fig. 1. Frequency of respiratory interventions for patients based on level of care.
Adenotonsillectomy is the surgical treatment of choice for uncomplicated children with OSA. The Childhood Adenotonsillectomy Trial was the first randomized study to demonstrate that AT significantly improves polysomnographic findings compared to watchful waiting in children aged 5–9 [7]. Patients with OSA are at a higher risk for postoperative respiratory compromise, a serious complication following AT [8]. The rate of this complication increases further in patients who are classified as high-risk, such as those with age < 3 years, weight < 5th percentile for age, craniofacial abnormalities, severe OSA, and obesity [8–10]. These patients are often electively admitted to the ICU for postoperative monitoring, but extant literature indicates that ICU care may be unnecessary for many of these patients [8,11–15]. Given the higher cost and resource demands associated with intensive care, a number of groups have attempted to identify which patients undergoing AT require an inpatient and/or ICU admission. Baguley et al. [11] recommended overnight admission for children with mild-moderate OSA who are < 2 years of age or have a significant comorbidity, postoperative oxygen desaturation < 90% on room air, or postoperative oxygen requirement > 6 h. Among 122 patients with severe OSA, Walker et al. [15] identified risk factors that defined their institution's PICU admission criteria: age < 2 years, respiratory disturbance index (RDI) > 24 events/hr, intraoperative laryngospasm requiring intervention, postoperative oxygen desaturation < 90% on room air, and total PACU stay > 100 min. Blenke et al. [12] recommended PICU referral for children with severe OSA and neurologic comorbidities, craniofacial abnormalities, morbid obesity, respiratory failure, or severe cyanotic congenital heart disease. Theilhaber et al. [8] reported their institution's PICU admission criteria to include age < 2 years, weight < 3rd percentile, morbid obesity, neuromuscular disease, severe OSA, and cyanotic heart disease. However, this group also discussed an alternative of monitoring oxygen saturation in the PACU or step-down unit to safely identify patients who may experience a severe event and ultimately require PICU care. In light of the significant
Fig. 2. Frequency of respiratory events and interventions for all patients, based on location of event or intervention.
seven unplanned PICU admissions experienced at least one severe respiratory event, while the other three had desaturations severe enough to necessitate escalation of care. One of these patients developed hypercarbic/hypoxic respiratory compromise after bronchoscopy and required multiple doses of albuterol, epinephrine, and 100% FiO2. This patient was directly admitted to the PICU from the operating room (OR) while still intubated. Comparison of non-PICU patients with unscheduled versus scheduled hospitalizations revealed a significantly greater length of stay (3.6 vs. 1.2 days, p < 0.001) and total number of days requiring supplemental oxygen above baseline (0.4 vs. 0.1 days, p < 0.001). There were no other significant differences between these groups. Comparing the unscheduled versus scheduled PICU admissions revealed no
Fig. 3. Depiction of respiratory events and associated escalations of care (n indicates number of patients).
34
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
Table 4 Comparison of pediatric intensive care unit and non-pediatric intensive care unit, scheduled vs. escalated admissions. PICU
Age (yr) Weight percentile-for-age (%) Comorbidities/patient Pre-op PSG (AHI) Admission length (days) Days requiring supplemental O2 (days) Percent days requiring O2 (%) Days requiring other respiratory intervention (days) Percent days requiring other respiratory intervention (%) Total PACU time (min) Total PACU time spent on supplemental O2 (min) Percent PACU time spent on supplemental O2 (%) Total# mild events Total# severe events
Non-PICU
Scheduled n = 29
Escalated n=7
p value
Scheduled n = 90
Escalated n=7
p value
3.1 ± 2.5 43.4 ± 34.7 1.6 ± 1.2 19.4 ± 17.5 2.9 ± 2.6 1.2 ± 1.9 32.9 ± 29.4 1.1 ± 1.5 31.5 ± 32.2 144.2 ± 86.3 98.0 ± 90.8 68.5 ± 38.3 27 7
4.7 ± 2.4 45.3 ± 43.8 2.1 ± 1.4 6.1 ± 4.8 4.7 ± 2.8 3.5 ± 2.7 63.1 ± 34.7 1.6 ± 1.2 50.9 ± 39.1 225.3 ± 121.3 176.2 ± 133.5 76.8 ± 38.6 5 4
0.136 0.896 0.292 0.056 0.103 0.009 0.459 0.025 0.179 0.047 0.071 0.611 – –
5.9 ± 4.1 56.1 ± 36.6 0.87 ± 1.08 6.0 ± 4.7 1.2 ± 1.3 0.1 ± 0.4 8.7 ± 26.0 0.1 ± 0.6 4.5 ± 19.0 146.4 ± 122.7 43.8 ± 58.5 30.4 ± 29.5 23 3
7.8 ± 5.1 39.9 ± 35.6 0.71 ± 0.95 4.9 ± 3.9 3.6 ± 1.9 0.4 ± 1.1 7.1 ± 18.9 0.0 ± 0.0 0.0 ± 0.0 112.4 ± 45.9 28.4 ± 36.8 23.9 ± 27.5 1 1
0.258 0.262 0.719 0.547 < 0.001 < 0.001 0.603 0.872 0.532 0.470 0.494 0.577 – –
P values reflect comparisons between the scheduled and escalated care groups, either in the PICU or non-PICU groups. Scheduled = scheduled admission; Escalated = unanticipated escalation of care.
escalation among all patients, as only 2 of the 7 unplanned non-PICU hospital stays were related to adverse postoperative respiratory events. Nevertheless, these 2 patients experienced adverse respiratory events in the PACU, supporting the utility of the postoperative PACU course as a potential indicator of patients' risk for respiratory compromise. Based on our results, the length of time spent in the PACU and the PACU time requiring supplemental oxygen may serve as specific markers of potential postoperative respiratory complications and patients' need for higher levels of care. This study further analyzed the relationship between severity of respiratory events and care intensity. Events of any severity occurred in 44.4% of patients. This rate is higher than rates quoted by Theilhaber (36.1%) and Blenke (24.1%), although they had smaller sample sizes of 26 and 49 patients, respectively [8,12]. Desaturations requiring re-initiation of supplemental oxygen accounted for almost two-thirds of all respiratory events in our cohort and most commonly occurred in the PACU, consistent with Theilhaber's findings [8]. Fifteen total severe events occurred, five of which led to an escalation of care. Only 31% of PICU patients (24% of planned PICU admissions) experienced a severe event, and 58.3% experienced only mild events. This low rate of severe events among PICU patients, along with the ability to manage mild respiratory events outside the ICU, suggests that many of our patients may be unnecessarily admitted to the PICU. Analysis of these severe respiratory events revealed a greater incidence in patients who had: 1) lower weight percentile-for-age (though this trend was not significant), in agreement with the literature [8,11–13,15], and 2) greater percentage of total PACU time and hospital days spent on supplemental oxygen. This latter finding further supports PACU time spent on supplemental oxygen as a potentially useful predictor of patients' required level of care. This study is limited by a number of factors. As a retrospective study, the data is limited by information recorded in electronic charts. Some patients' PACU records were missing or lacked clear details of the postoperative course. Our small sample size may have limited our ability to detect statistically significant differences. Some patients remained in the PACU for reasons other than respiratory complications, which may have created a positive bias for total time spent in the PACU. Furthermore, details about only the most severe respiratory event were recorded for each patient, so comprehensive information about other respiratory events that occurred during the postoperative hospital course may be missing. Despite these limitations, this study provides data that will inform future work to develop an algorithm that can accurately predict the need for ICU stay after AT in patients with OSA. This algorithm will be trialed in a prospective study, which may further
variation in admission criteria between institutions and the continuing evolution of institutional standards, a more sensitive and specific method of predicting possible respiratory complications and patient need for ICU admission is needed. At our institution, determination of postoperative ICU admission is not strictly defined and has significant variation by provider. The criteria considered include: significant comorbidity (including Down syndrome, Pierre Robin sequence, cerebral palsy, and neurodevelopmental or neuromuscular disease); severe respiratory event in the OR or PACU (e.g., laryngospasm); oxygen saturation < 90% on room air in the PACU; oxygen requirement in the PACU exceeding 40% FiO2 by facemask; age < 24 months; and AHI > 24 [16]. In our study of 133 patients, the 36 patients admitted to the PICU were younger, had higher preoperative AHIs, and had more comorbid risk factors for respiratory compromise. These results agree with cited risk factors for postoperative respiratory complications and with the aforementioned studies detailing various ICU admission criteria [8,11–15]. During their postoperative hospital course, PICU patients spent a greater percentage of PACU time on supplemental oxygen and required additional respiratory interventions on more hospital days. To our knowledge, both of these findings have not been previously reported. Similar to Walker et al.'s study, our PICU cohort also spent an average time in the PACU > 100 min, but this was not significantly greater than non-PICU patients [15]. Walker et al.'s study included only patients with severe OSA, and 41% of patients had a comorbid risk factor. While the preoperative AHI across all our patients was just below the threshold for severe OSA of 10.0 events/hour, this measure precludes comparison to the RDI-based measure of OSA severity in the Walker et al. study. Nevertheless, this discrepancy may be explained by the fact that 57% of our patients had a comorbid risk factor; our patient cohort may have been a higher-risk group that required more care in the PACU. Further analysis revealed that 14 patients had an unscheduled escalation of care. Age, weight percentile-for-age, and comorbidities per patient were not identified as risk factors. Among the 7 patients with unscheduled PICU admissions, compared to patients with scheduled PICU admissions, there was a trend towards lower preoperative AHIs. This may be explained by our providers' elective postoperative PICU admission of patients with severe OSA. This result also suggests that patients requiring escalation of care to the ICU may have other more significant risk factors for respiratory compromise. Our results demonstrated that all 14 patients requiring any escalation of care spent more days on supplemental oxygen above baseline requirements and generally required longer PACU stays and longer PACU time on supplemental oxygen. The PACU course was not predictive of need for 35
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
and through the Section for Surgical Sciences at Vanderbilt University Medical Center. The research utilized REDCap which was funded by grant UL1TR000445 from NCATS/NIH.
identify other potential factors that can predict the need for an ICU level of care, such as intraoperative course or pain control modality. 5. Conclusion
References This study examines post-AT respiratory complications in pediatric OSA patients in order to identify possible factors that may correlate with adverse postoperative respiratory events and necessity for an ICU level of care. Our study identified important variables associated with PICU admission, escalation of care, and severe respiratory events. These variables include younger age, higher preoperative AHI, greater number of comorbid risk factors for respiratory compromise, more total time spent in the PACU, greater percentage of PACU time spent on supplemental oxygen, and more hospital days requiring oxygen above baseline or interventions besides supplemental oxygen. Future work will involve incorporation of these variables into a best-practice algorithm, which will stratify patients into risk groups for postoperative respiratory compromise in order to improve allocation of healthcare resources.
[1] [2] [3] [4] [5]
[6]
[7]
[8] [9]
Funding
[10]
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
[11] [12]
Conflicts of interest [13]
The authors report no conflicts of interest. [14]
Acknowledgements
[15]
This research was supported in part by the Surgical Outcomes Center for Kids at Monroe Carell Jr. Children's Hospital at Vanderbilt
[16]
36
Z. Ehsan, S.L. Ishman, Otolaryngol. Clin. 2016 (49) (2016) 1449–1464. M. Kohler, Curr. Opin. Allergy Clin. Immunol. 2009 (9) (2009) 4–9. J.L. Mathew, I. Narang, Paediatr. Respir. Rev. 2014 (15) (2014) 211–218. H.L. Tan, D. Gozal, L. Kheirandish-Gozal, Nat. Sci. Sleep 2013 (5) (2013) 109–123. 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, R.H. Schwartz, G. Setzen, E.R. Wald, E. Wall, G. Sandberg, M.M. PatelAmerican Academy of O-H, Neck Surgery F, Otolaryngol. Head Neck Surg. 2011 (144) (2011) S1–S30. C.L. Marcus, L.J. Brooks, K.A. Draper, D. Gozal, A.C. Halbower, J. Jones, M.S. Schechter, S.D. Ward, S.H. Sheldon, R.N. Shiffman, C. Lehmann, K. SpruytAmerican Academy of P, Pediatrics 2012 (130) (2012) e714–755. C.L. Marcus, R.H. Moore, C.L. Rosen, B. Giordani, S.L. Garetz, H.G. Taylor, R.B. Mitchell, R. Amin, E.S. Katz, R. Arens, S. Paruthi, H. Muzumdar, D. Gozal, N.H. Thomas, J. Ware, D. Beebe, K. Snyder, L. Elden, R.C. Sprecher, P. Willging, D. Jones, J.P. Bent, T. Hoban, R.D. Chervin, S.S. Ellenberg, S. RedlineChildhood Adenotonsillectomy T, N. Engl. J. Med. 2013 (368) (2013) 2366–2376. M. Theilhaber, S. Arachchi, D.S. Armstrong, M.J. Davey, G.M. Nixon, Int. J. Pediatr. Otorhinolaryngol. 2014 (78) (2014) 744–747. S.A. McColley, M.M. April, J.L. Carroll, R.M. Naclerio, G.M. Loughlin, Arch. Otolaryngol. Head Neck Surg. 1992 (118) (1992) 940–943. G.M. Rosen, R.P. Muckle, M.W. Mahowald, G.S. Goding, C. Ullevig, Pediatrics 1994 (93) (1994) 784–788. K.E. Baguley, A.T. Cheng, C. Castro, N. Wainbergas, K.A. Waters, Int. J. Pediatr. Otorhinolaryngol. 2014 (78) (2014) 71–74. E.J. Blenke, A.R. Anderson, H. Raja, S. Bew, L.C. Knight, J. Laryngol. Otol. 2008 (122) (2008) 42–45. G. del-Rio Camacho, M. Martinez Gonzalez, J. Sanabria Brossart, E. Gutierrez Moreno, T. Gomez Garcia, F. Troncoso Acevedo, Acta Otorrinolaringol. Esp. 2014 (65) (2014) 302–307. N.P. Shine, H.L. Coates, F.J. Lannigan, A.W. Duncan, Anaesth. Intensive Care 2006 (34) (2006) 724–730. P. Walker, B. Whitehead, M. Rowley, J. Laryngol. Otol. 127 (Suppl 1) (2013) S26–S29 2013. C.A. Hill, A. Litvak, C. Canapari, B. Cummings, C. Collins, D.G. Keamy, T.G. Ferris, C.J. Hartnick, Int. J. Pediatr. Otorhinolaryngol. 2011 (75) (2011) 1385–1390.
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Respiratory events after adenotonsillectomy requiring escalated admission status in children with obstructive sleep apnea
T
Alexandra M. Arambulaa,b, Deborah X. Xiea,b, Amy S. Whighama,b,c,∗ a
Vanderbilt University School of Medicine, Nashville, TN USA Surgical Outcomes Center for Kids, Vanderbilt University Medical Center, Nashville, TN, USA c Department of Otolaryngology - Head and Neck Surgery, Pediatric Otolaryngology, Vanderbilt University Medical Center, Nashville, TN USA b
A R T I C L E I N F O
A B S T R A C T
Keywords: Adenotonsillectomy Tonsillectomy Obstructive sleep apnea Postoperative complications Intensive care Pediatric
Objectives: To characterize postoperative respiratory complications following adenotonsillectomy (AT) in children with obstructive sleep apnea (OSA) and to identify variables associated with pediatric intensive care unit (PICU) admission. Methods: Retrospective analysis of 133 pediatric OSA patients with prior AT. Assessment of the postoperative hospital course informed patient stratification based on respiratory event severity, PICU admission status, and unscheduled escalation of care. Results: Thirty-six (26.8%) patients were admitted to the PICU. Compared to non-PICU admissions, these patients were significantly younger and with greater preoperative apnea-hypopnea indices, comorbidities, and percentage of post-anesthesia care unit (PACU) time requiring supplemental oxygen. Seventy-one respiratory events occurred in 59 patients, with 60.6% affecting PICU patients. Fifteen severe events occurred, affecting 31% of PICU patients. Of 14 unscheduled escalations of care, 7 were PICU admissions who, compared to planned PICU admissions, spent significantly more time in the PACU and exhibited a trend towards greater PACU time on supplemental oxygen. Conclusions: Pediatric patients requiring post-AT PICU care have more risk factors for respiratory compromise. Total PACU time and total PACU time requiring supplemental oxygen may indicate patient risk for postoperative respiratory complications and need for intensive care. Future work includes prospective determination of appropriate post-AT PICU admission.
1. Introduction Obstructive sleep apnea (OSA) is an increasingly common sleeprelated breathing disorder that affects 1–5% of children [1]. OSA is characterized by disruptions in sleep patterns secondary to partial or complete collapse of the upper airway. Well-known risk factors for OSA include male gender, black race, obesity, craniofacial disorders, neuromuscular disease, and Down syndrome [1–4]. As adenotonsillar hypertrophy (ATH) contributes to a large percentage of pediatric OSA cases, the American Academy of Pediatrics and American Academy of Otolaryngology – Head and Neck Surgery recommend adenotonsillectomy (AT) as the primary treatment modality for children with OSA and ATH, as this procedure leads to significantly improved polysomnographic findings in uncomplicated patients [1,5–7]. One serious complication associated with AT is postoperative respiratory compromise secondary to upper airway obstruction. This risk increases from 0-1.3% to 16–27% with a diagnosis of OSA [8]. The
∗
likelihood of respiratory compromise further increases in high-risk patients, including children with age < 3 years, weight < 5th percentile for age, craniofacial abnormalities, severe OSA as measured on polysomnography (PSG), cardiac disease, prematurity, obesity, and hypotonia [8–10]. Identification of these characteristics in the early 1990s led many hospitals to electively admit high-risk patients to the pediatric intensive care unit (PICU) for airway monitoring after AT. Retrospective reviews indicate that routine postoperative ICU admission may be unnecessary for many of these patients, although no consensus exists [8,11–15]. Our own institution lacks clear guidelines regarding postoperative elective ICU admission. Given the increased financial cost of ICU admission and limited availability of PICU beds, identification of patient variables associated with a need for intensive care is essential in developing a best-practice algorithm. This study investigates post-AT respiratory complications in pediatric OSA patients in order to identify possible factors that may correlate with necessity for an ICU level of care.
Corresponding author. Doctors' Office Tower, 7th Floor, 2200 Children's Way, Nashville, TN 37232-8605, USA. E-mail address: [email protected] (A.S. Whigham).
https://doi.org/10.1016/j.ijporl.2018.01.009 Received 27 September 2017; Received in revised form 10 January 2018; Accepted 14 January 2018 Available online 31 January 2018 0165-5876/ © 2018 Elsevier B.V. All rights reserved.
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
2. Methods
3. Results
This retrospective review was approved by the Vanderbilt University Institutional Review Board (IRB #170372). Patients with diagnosed sleep disordered breathing (SDB) or OSA (clinically or by PSG) managed by the pediatric otolaryngology service at our institution were identified from a clinic database. Inclusion criteria were patients less than 18 years of age, a diagnosis of sleep disordered breathing (SDB) or OSA (clinically or by PSG), evaluation by otolaryngologists at our institution between 2014 and 2015, and prior AT at our institution. Patient demographic information was recorded. Comorbid risk factors were defined as Down syndrome, cerebral palsy, hypotonia, craniofacial abnormalities, neuromuscular disease, or laryngotracheomalacia. Data about each patient's postoperative hospital course were also collected. Though surgeon and anesthesiologist dependent, this institution's postoperative ICU admission criteria include factors identified by Hill et al.: obstructive apnea-hypopnea index (AHI) > 24; age < 24 months; oxygen saturation < 90% on room air in the post-anesthesia care unit (PACU); and intraoperative complications including bronchospasm or laryngospasm requiring treatment [16]. Additional admission criteria include airway obstruction requiring nasal trumpet placement; oxygen requirement exceeding 40% fraction of inspired oxygen (FiO2) in the PACU; craniofacial abnormalities including Down syndrome and Pierre Robin sequence; and neuromuscular/neurodevelopmental disorders such as cerebral palsy and other conditions where patients require frequent suctioning to assist in management of secretions. Respiratory events occurring during the postoperative hospital course, defined as the time period from surgical extubation to discharge, were recorded and stratified into mild and severe events (Table 1). A desaturation requiring multiple interventions was counted as a single respiratory event. Desaturations requiring interventions under both the mild and severe event categories were classified as severe respiratory events. Respiratory events were further categorized based on whether the complications required an escalation of care. This was defined as any unplanned admission to either the hospital floor or PICU. Patients experiencing multiple respiratory events were categorized based on the most severe event. Descriptive statistics were performed and computed as means with standard deviation (SD) for quantitative variables and counts and percentages for categorical variables. Differences between patients who did or did not require PICU admission and between patients who did or did not have an unexpected escalation of care were determined. Unpaired, two-sample t-tests and univariate analysis were used to compare continuous variables, and chi-squared tests were used to compare categorical variables. Statistical significance was set a priori at p < 0.05.
Of 312 pediatric patients with SDB/OSA, 133 met inclusion criteria. Demographic information is shown in Table 2. A small majority of patients were male (56.4%), with this predominance increasing further among PICU patients (69.4%). Over 90% of all patients identified as either white or African American. Preoperative PSG was performed in 105 (78.9%) patients, with obstructive AHIs ranging from 0.6 to 63.5 events/hr. Fifty-seven percent of all patients had a comorbid diagnosis (47% of standard admission [non-PICU] patients, 83% of PICU patients). Following their operation, two patients were directly admitted to the PICU. The remaining 131 patients went to the PACU and were either discharged home (20 patients, 15.3%) or admitted to the hospital floor (77 patients, 58.8%) or the PICU (34 patients, 26.0%). Thus, a total of 36 patients (27.1%) were admitted to the PICU. As seen in Table 2, these patients were significantly younger than non-PICU patients, with higher preoperative AHIs, and with more comorbid risk factors for respiratory compromise. Table 3 contains information on the patients' immediate postoperative and hospital course. Within this group, 7 patients (1 PICU, 6 non-PICU) had missing PACU records. The two groups spent a similar amount of time in the PACU; however, patients admitted to the PICU ultimately spent a significantly longer percentage of this time on supplemental oxygen. They also had significantly longer hospital admissions (p < 0.001), more hospital days on supplemental oxygen (p < 0.001), and more hospital days requiring additional respiratory support (p = 0.041). Seventy-one respiratory events occurred in 59 patients (44.4%), requiring 125 interventions (Table 3). The most common respiratory event was a desaturation (n = 55, 77.5% of all events). The average oxygen saturation for children placed on supplemental oxygen was 82.9 ± 9.8%, with a trend towards lower saturations in the PICU group (Table 3). The lowest recorded saturations were 53% and 50% in the PACU and PICU, respectively. Nearly all desaturations were treated with administration of supplemental oxygen (n = 54, 43.2% of all interventions, Fig. 1). Of all events requiring supplemental oxygen, 61.1% occurred in the PACU (Fig. 2). Additional respiratory interventions were more commonly needed by patients ultimately admitted to the PICU (63 vs. 8 in non-PICU patients). These interventions more often occurred in the PICU compared to the PACU, including placement of a nasal trumpet (n = 13, 46.2% vs. 30.8%), airway suctioning (n = 19, 47.4% vs. 31.6%), and steroid administration (n = 11, 45.5% vs. 27.3%). Fifteen respiratory events (21.1%) were classified as severe, occurring in 15 patients (11 PICU, 4 non-PICU). The remaining 56 mild respiratory events (78.9%) occurred in 50 patients, including 6 of the
Table 1 Stratification of respiratory events. Mild respiratory events
Apnea Stridor Laryngospasm Bronchospasm Oxygen desaturation requiring at least one of the following: Re-initiation of supplemental oxygen Jaw thrust Nasal trumpet placement Airway suctioning
Severe respiratory events
Respiratory distress requiring initiation of rapid response team Oxygen desaturation requiring at least one of the following: Bronchodilator Muscle relaxant Initiation of continuous positive airway pressure (CPAP) Initiation of bilevel positive airway pressure (BiPAP) Intubation
32
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
Table 2 Demographic and clinical patient data.
Age at surgery (years) Sex Female Male Raceb White African-American Asian Hispanic Native Hawaiian/Other Pacific Islander Weight percentile-for-age (%) Pre-AT AHI (events/hr) Number of comorbidities (per patient)
All Patients (n = 133), n (%); mean ± SDa
Non-PICU (n = 97), n (%); mean ± SDa
PICU (n = 36), n (%); mean ± SDa
p value
5.3 ± 4.0
6.1 ± 4.2
3.4 ± 2.5
< 0.001 0.064
58 (43.6%) 75 (56.4%)
47 (48.5%) 50 (51.5%)
11 (30.6%) 25 (69.4%)
90 (67.7%) 32 (24.1%) 3 (2.3%) 6 (4.5%) 1 (0.7%)
69 (71.1%) 19 (19.6%) 3 (3.1%) 5 (5.2%) 1 (1.0%)
21 (58.3%) 13 (36.1%) – 1 (2.8%) –
51.9 ± 36.6 9.4 ± 11.3 1.1 ± 1.2
54.9 ± 36.6 6.0 ± 4.6 0.9 ± 1.1
43.8 ± 35.9 17.3 ± 16.8 1.7 ± 1.2
< 0.001 0.161 0.048 0.557
0.120 < 0.001 < 0.001
Data for all patients, further subdivided by non-PICU or PICU admission status. Non-PICU patients were children requiring only standard admission. PICU patients were children requiring admission to the PICU. P values reflect comparisons between the PICU and non-PICU groups. a Categorical variables are described as n (%); continuous variables are described as mean ± SD. b 1 PICU patient declined to indicate race.
floor rather than same-day discharge, 1 to the PICU rather than sameday discharge, and 2 to the PICU rather than scheduled floor admission. Five of the 15 patients with severe respiratory events (33.3%) required an escalation of care: 1 floor admission rather than same-day discharge and 4 PICU admissions rather than hospital floor admissions (Fig. 3). In total, 14 patients had an unscheduled escalation of care (Table 4). Seven required unplanned PICU admission (19.4% of PICU patients). Seven had an unplanned hospital admission (7.2% of non-PICU patients), but only 2 were admitted for respiratory concerns, accounting for one severe and one mild respiratory event. Other reasons for admission included poor oral intake, nausea and vomiting. Four of the
patients who experienced a severe event. Compared to those experiencing mild respiratory events, patients experiencing severe events had a lower weight percentile-for-age (29.7% vs. 49.1%, p = 0.057). There was a statistically significant difference in length of admission (4.5 vs. 2.1 days, p < 0.001), percent of hospital days requiring supplemental oxygen (60.7% vs. 23.6%, p < 0.001), and percent of total PACU time requiring supplemental oxygen (84% vs. 58%, p = 0.026) between these groups. There were no significant differences based on age, preoperative PSG, comorbid risk factors per patient, or total time spent in the PACU. Of the patients who experienced one or more mild respiratory events, 4 (8.0%) needed an escalation of care: 1 to the hospital
Table 3 Postoperative respiratory events/interventions and oxygen requirements.
Total time in PACU (min) Lowest O2 saturation prior to supplemental O2 administration (%) Total PACU time on O2 (min) % total PACU time on O2 (%) Number of respiratory events Apnea Stridor Oxygen desaturation Laryngospasm Bronchospasm Number of interventionsb Nasal trumpet Jaw thrust Suction Postoperative oxygen Muscle relaxants Bronchodilators Steroids CPAP BiPAP Rapid response Intubation Length of hospital admission (days) [range; median] Proportion of admission days requiring supplemental O2 [range; median] Proportion of admission days requiring other respiratory interventions [range; median]
Non-PICU (n = 97); n (%); mean ± SDa
PICU (n = 36); n (%); mean ± SDa
p value
144.5 ± 119.9 84.6 ± 8.4 43.0 ± 57.5 30.1 ± 29.3 28 (39.4%) – 2 (7.1%) 26 (36.6%) – – 34 (27.2%) 1 (2.9%) – 2 (5.9%) 26 (76.5%) – 3 (8.8%) 1 (2.9%) 1 (2.9%) – – – 1.3 ± 1.4 [0–8; 1] 0.09 ± 0.25 [0–3; 0] 0.04 ± 0.35 [0–4; 0]
158.1 ± 96.2 81.2 ± 10.9 111.4 ± 101.5 70.0 ± 37.9 43 (60.6%) 1 (1.4%) 11 (15.5%) 29 (40.8%) 2 (2.8%) – 91 (73.6%) 12 (13.2%) 5 (5.5%) 17 (18.7%) 28 (30.8%) 2 (2.2%) 5 (5.5%) 10 (11.0%) 5 (5.5%) 2 (2.2%) 2 (2.2%) 3 (3.3%) 3.2 ± 2.7 [1–11; 2] 0.39 ± 0.32 [0–9; 0.625] 0.18 ± 0.34 [0–5; 1]
0.542 0.058 < 0.001 < 0.001
< 0.001 < 0.001 0.041
Clinical data from the immediate postoperative course and entire hospital admission. All percentages are calculated based on total patient population, unless otherwise specified. P values reflect comparisons between the PICU and non-PICU groups. O2 = supplemental oxygen. a Categorical variables are described as n (%); continuous variables are described as mean ± SD. b Percentages for intervention subcategories were calculated from the number of interventions in respective patient groups (non-PICU or PICU).
33
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
statistically significant difference in demographic or clinical characteristics. There was a trend towards greater pre-operative AHIs for patients with scheduled PICU admission (p = 0.056). The postoperative course differed significantly for days on supplemental oxygen above baseline requirements (3.5 vs. 1.2 days, p = 0.009), days requiring additional respiratory interventions (1.6 vs. 1.1 days, p = 0.025), and total time spent in the PACU (225 vs. 144 min, p = 0.047). There was a trend towards greater total PACU time spent on supplemental oxygen (176 vs. 98 min, p = 0.071) in the group with unscheduled admissions, but there was no difference in number of severe events per patient (0.24 [median 0] vs. 0.57 [median 1] events, p = 0.294). 4. Discussion
Fig. 1. Frequency of respiratory interventions for patients based on level of care.
Adenotonsillectomy is the surgical treatment of choice for uncomplicated children with OSA. The Childhood Adenotonsillectomy Trial was the first randomized study to demonstrate that AT significantly improves polysomnographic findings compared to watchful waiting in children aged 5–9 [7]. Patients with OSA are at a higher risk for postoperative respiratory compromise, a serious complication following AT [8]. The rate of this complication increases further in patients who are classified as high-risk, such as those with age < 3 years, weight < 5th percentile for age, craniofacial abnormalities, severe OSA, and obesity [8–10]. These patients are often electively admitted to the ICU for postoperative monitoring, but extant literature indicates that ICU care may be unnecessary for many of these patients [8,11–15]. Given the higher cost and resource demands associated with intensive care, a number of groups have attempted to identify which patients undergoing AT require an inpatient and/or ICU admission. Baguley et al. [11] recommended overnight admission for children with mild-moderate OSA who are < 2 years of age or have a significant comorbidity, postoperative oxygen desaturation < 90% on room air, or postoperative oxygen requirement > 6 h. Among 122 patients with severe OSA, Walker et al. [15] identified risk factors that defined their institution's PICU admission criteria: age < 2 years, respiratory disturbance index (RDI) > 24 events/hr, intraoperative laryngospasm requiring intervention, postoperative oxygen desaturation < 90% on room air, and total PACU stay > 100 min. Blenke et al. [12] recommended PICU referral for children with severe OSA and neurologic comorbidities, craniofacial abnormalities, morbid obesity, respiratory failure, or severe cyanotic congenital heart disease. Theilhaber et al. [8] reported their institution's PICU admission criteria to include age < 2 years, weight < 3rd percentile, morbid obesity, neuromuscular disease, severe OSA, and cyanotic heart disease. However, this group also discussed an alternative of monitoring oxygen saturation in the PACU or step-down unit to safely identify patients who may experience a severe event and ultimately require PICU care. In light of the significant
Fig. 2. Frequency of respiratory events and interventions for all patients, based on location of event or intervention.
seven unplanned PICU admissions experienced at least one severe respiratory event, while the other three had desaturations severe enough to necessitate escalation of care. One of these patients developed hypercarbic/hypoxic respiratory compromise after bronchoscopy and required multiple doses of albuterol, epinephrine, and 100% FiO2. This patient was directly admitted to the PICU from the operating room (OR) while still intubated. Comparison of non-PICU patients with unscheduled versus scheduled hospitalizations revealed a significantly greater length of stay (3.6 vs. 1.2 days, p < 0.001) and total number of days requiring supplemental oxygen above baseline (0.4 vs. 0.1 days, p < 0.001). There were no other significant differences between these groups. Comparing the unscheduled versus scheduled PICU admissions revealed no
Fig. 3. Depiction of respiratory events and associated escalations of care (n indicates number of patients).
34
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
Table 4 Comparison of pediatric intensive care unit and non-pediatric intensive care unit, scheduled vs. escalated admissions. PICU
Age (yr) Weight percentile-for-age (%) Comorbidities/patient Pre-op PSG (AHI) Admission length (days) Days requiring supplemental O2 (days) Percent days requiring O2 (%) Days requiring other respiratory intervention (days) Percent days requiring other respiratory intervention (%) Total PACU time (min) Total PACU time spent on supplemental O2 (min) Percent PACU time spent on supplemental O2 (%) Total# mild events Total# severe events
Non-PICU
Scheduled n = 29
Escalated n=7
p value
Scheduled n = 90
Escalated n=7
p value
3.1 ± 2.5 43.4 ± 34.7 1.6 ± 1.2 19.4 ± 17.5 2.9 ± 2.6 1.2 ± 1.9 32.9 ± 29.4 1.1 ± 1.5 31.5 ± 32.2 144.2 ± 86.3 98.0 ± 90.8 68.5 ± 38.3 27 7
4.7 ± 2.4 45.3 ± 43.8 2.1 ± 1.4 6.1 ± 4.8 4.7 ± 2.8 3.5 ± 2.7 63.1 ± 34.7 1.6 ± 1.2 50.9 ± 39.1 225.3 ± 121.3 176.2 ± 133.5 76.8 ± 38.6 5 4
0.136 0.896 0.292 0.056 0.103 0.009 0.459 0.025 0.179 0.047 0.071 0.611 – –
5.9 ± 4.1 56.1 ± 36.6 0.87 ± 1.08 6.0 ± 4.7 1.2 ± 1.3 0.1 ± 0.4 8.7 ± 26.0 0.1 ± 0.6 4.5 ± 19.0 146.4 ± 122.7 43.8 ± 58.5 30.4 ± 29.5 23 3
7.8 ± 5.1 39.9 ± 35.6 0.71 ± 0.95 4.9 ± 3.9 3.6 ± 1.9 0.4 ± 1.1 7.1 ± 18.9 0.0 ± 0.0 0.0 ± 0.0 112.4 ± 45.9 28.4 ± 36.8 23.9 ± 27.5 1 1
0.258 0.262 0.719 0.547 < 0.001 < 0.001 0.603 0.872 0.532 0.470 0.494 0.577 – –
P values reflect comparisons between the scheduled and escalated care groups, either in the PICU or non-PICU groups. Scheduled = scheduled admission; Escalated = unanticipated escalation of care.
escalation among all patients, as only 2 of the 7 unplanned non-PICU hospital stays were related to adverse postoperative respiratory events. Nevertheless, these 2 patients experienced adverse respiratory events in the PACU, supporting the utility of the postoperative PACU course as a potential indicator of patients' risk for respiratory compromise. Based on our results, the length of time spent in the PACU and the PACU time requiring supplemental oxygen may serve as specific markers of potential postoperative respiratory complications and patients' need for higher levels of care. This study further analyzed the relationship between severity of respiratory events and care intensity. Events of any severity occurred in 44.4% of patients. This rate is higher than rates quoted by Theilhaber (36.1%) and Blenke (24.1%), although they had smaller sample sizes of 26 and 49 patients, respectively [8,12]. Desaturations requiring re-initiation of supplemental oxygen accounted for almost two-thirds of all respiratory events in our cohort and most commonly occurred in the PACU, consistent with Theilhaber's findings [8]. Fifteen total severe events occurred, five of which led to an escalation of care. Only 31% of PICU patients (24% of planned PICU admissions) experienced a severe event, and 58.3% experienced only mild events. This low rate of severe events among PICU patients, along with the ability to manage mild respiratory events outside the ICU, suggests that many of our patients may be unnecessarily admitted to the PICU. Analysis of these severe respiratory events revealed a greater incidence in patients who had: 1) lower weight percentile-for-age (though this trend was not significant), in agreement with the literature [8,11–13,15], and 2) greater percentage of total PACU time and hospital days spent on supplemental oxygen. This latter finding further supports PACU time spent on supplemental oxygen as a potentially useful predictor of patients' required level of care. This study is limited by a number of factors. As a retrospective study, the data is limited by information recorded in electronic charts. Some patients' PACU records were missing or lacked clear details of the postoperative course. Our small sample size may have limited our ability to detect statistically significant differences. Some patients remained in the PACU for reasons other than respiratory complications, which may have created a positive bias for total time spent in the PACU. Furthermore, details about only the most severe respiratory event were recorded for each patient, so comprehensive information about other respiratory events that occurred during the postoperative hospital course may be missing. Despite these limitations, this study provides data that will inform future work to develop an algorithm that can accurately predict the need for ICU stay after AT in patients with OSA. This algorithm will be trialed in a prospective study, which may further
variation in admission criteria between institutions and the continuing evolution of institutional standards, a more sensitive and specific method of predicting possible respiratory complications and patient need for ICU admission is needed. At our institution, determination of postoperative ICU admission is not strictly defined and has significant variation by provider. The criteria considered include: significant comorbidity (including Down syndrome, Pierre Robin sequence, cerebral palsy, and neurodevelopmental or neuromuscular disease); severe respiratory event in the OR or PACU (e.g., laryngospasm); oxygen saturation < 90% on room air in the PACU; oxygen requirement in the PACU exceeding 40% FiO2 by facemask; age < 24 months; and AHI > 24 [16]. In our study of 133 patients, the 36 patients admitted to the PICU were younger, had higher preoperative AHIs, and had more comorbid risk factors for respiratory compromise. These results agree with cited risk factors for postoperative respiratory complications and with the aforementioned studies detailing various ICU admission criteria [8,11–15]. During their postoperative hospital course, PICU patients spent a greater percentage of PACU time on supplemental oxygen and required additional respiratory interventions on more hospital days. To our knowledge, both of these findings have not been previously reported. Similar to Walker et al.'s study, our PICU cohort also spent an average time in the PACU > 100 min, but this was not significantly greater than non-PICU patients [15]. Walker et al.'s study included only patients with severe OSA, and 41% of patients had a comorbid risk factor. While the preoperative AHI across all our patients was just below the threshold for severe OSA of 10.0 events/hour, this measure precludes comparison to the RDI-based measure of OSA severity in the Walker et al. study. Nevertheless, this discrepancy may be explained by the fact that 57% of our patients had a comorbid risk factor; our patient cohort may have been a higher-risk group that required more care in the PACU. Further analysis revealed that 14 patients had an unscheduled escalation of care. Age, weight percentile-for-age, and comorbidities per patient were not identified as risk factors. Among the 7 patients with unscheduled PICU admissions, compared to patients with scheduled PICU admissions, there was a trend towards lower preoperative AHIs. This may be explained by our providers' elective postoperative PICU admission of patients with severe OSA. This result also suggests that patients requiring escalation of care to the ICU may have other more significant risk factors for respiratory compromise. Our results demonstrated that all 14 patients requiring any escalation of care spent more days on supplemental oxygen above baseline requirements and generally required longer PACU stays and longer PACU time on supplemental oxygen. The PACU course was not predictive of need for 35
International Journal of Pediatric Otorhinolaryngology 107 (2018) 31–36
A.M. Arambula et al.
and through the Section for Surgical Sciences at Vanderbilt University Medical Center. The research utilized REDCap which was funded by grant UL1TR000445 from NCATS/NIH.
identify other potential factors that can predict the need for an ICU level of care, such as intraoperative course or pain control modality. 5. Conclusion
References This study examines post-AT respiratory complications in pediatric OSA patients in order to identify possible factors that may correlate with adverse postoperative respiratory events and necessity for an ICU level of care. Our study identified important variables associated with PICU admission, escalation of care, and severe respiratory events. These variables include younger age, higher preoperative AHI, greater number of comorbid risk factors for respiratory compromise, more total time spent in the PACU, greater percentage of PACU time spent on supplemental oxygen, and more hospital days requiring oxygen above baseline or interventions besides supplemental oxygen. Future work will involve incorporation of these variables into a best-practice algorithm, which will stratify patients into risk groups for postoperative respiratory compromise in order to improve allocation of healthcare resources.
[1] [2] [3] [4] [5]
[6]
[7]
[8] [9]
Funding
[10]
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
[11] [12]
Conflicts of interest [13]
The authors report no conflicts of interest. [14]
Acknowledgements
[15]
This research was supported in part by the Surgical Outcomes Center for Kids at Monroe Carell Jr. Children's Hospital at Vanderbilt
[16]
36
Z. Ehsan, S.L. Ishman, Otolaryngol. Clin. 2016 (49) (2016) 1449–1464. M. Kohler, Curr. Opin. Allergy Clin. Immunol. 2009 (9) (2009) 4–9. J.L. Mathew, I. Narang, Paediatr. Respir. Rev. 2014 (15) (2014) 211–218. H.L. Tan, D. Gozal, L. Kheirandish-Gozal, Nat. Sci. Sleep 2013 (5) (2013) 109–123. 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, R.H. Schwartz, G. Setzen, E.R. Wald, E. Wall, G. Sandberg, M.M. PatelAmerican Academy of O-H, Neck Surgery F, Otolaryngol. Head Neck Surg. 2011 (144) (2011) S1–S30. C.L. Marcus, L.J. Brooks, K.A. Draper, D. Gozal, A.C. Halbower, J. Jones, M.S. Schechter, S.D. Ward, S.H. Sheldon, R.N. Shiffman, C. Lehmann, K. SpruytAmerican Academy of P, Pediatrics 2012 (130) (2012) e714–755. C.L. Marcus, R.H. Moore, C.L. Rosen, B. Giordani, S.L. Garetz, H.G. Taylor, R.B. Mitchell, R. Amin, E.S. Katz, R. Arens, S. Paruthi, H. Muzumdar, D. Gozal, N.H. Thomas, J. Ware, D. Beebe, K. Snyder, L. Elden, R.C. Sprecher, P. Willging, D. Jones, J.P. Bent, T. Hoban, R.D. Chervin, S.S. Ellenberg, S. RedlineChildhood Adenotonsillectomy T, N. Engl. J. Med. 2013 (368) (2013) 2366–2376. M. Theilhaber, S. Arachchi, D.S. Armstrong, M.J. Davey, G.M. Nixon, Int. J. Pediatr. Otorhinolaryngol. 2014 (78) (2014) 744–747. S.A. McColley, M.M. April, J.L. Carroll, R.M. Naclerio, G.M. Loughlin, Arch. Otolaryngol. Head Neck Surg. 1992 (118) (1992) 940–943. G.M. Rosen, R.P. Muckle, M.W. Mahowald, G.S. Goding, C. Ullevig, Pediatrics 1994 (93) (1994) 784–788. K.E. Baguley, A.T. Cheng, C. Castro, N. Wainbergas, K.A. Waters, Int. J. Pediatr. Otorhinolaryngol. 2014 (78) (2014) 71–74. E.J. Blenke, A.R. Anderson, H. Raja, S. Bew, L.C. Knight, J. Laryngol. Otol. 2008 (122) (2008) 42–45. G. del-Rio Camacho, M. Martinez Gonzalez, J. Sanabria Brossart, E. Gutierrez Moreno, T. Gomez Garcia, F. Troncoso Acevedo, Acta Otorrinolaringol. Esp. 2014 (65) (2014) 302–307. N.P. Shine, H.L. Coates, F.J. Lannigan, A.W. Duncan, Anaesth. Intensive Care 2006 (34) (2006) 724–730. P. Walker, B. Whitehead, M. Rowley, J. Laryngol. Otol. 127 (Suppl 1) (2013) S26–S29 2013. C.A. Hill, A. Litvak, C. Canapari, B. Cummings, C. Collins, D.G. Keamy, T.G. Ferris, C.J. Hartnick, Int. J. Pediatr. Otorhinolaryngol. 2011 (75) (2011) 1385–1390.