|Ahead of print
|Perioperative adverse respiratory events in children with obstructive sleep apnoea
Swapna Thampi1, Shang Yee Chong2, Dilip Kumar Pawar3
1 Department of Paediatric Anaesthesia, KK Women's and Children's Hospital; Department of Anaesthesia, National University Hospital, Singapore
2 Department of Paediatric Anaesthesia, KK Women's and Children's Hospital; Department of Anaesthesiology, Intensive Care and Pain Medicine, Tan Tock Seng Hospital, Singapore
3 Department of Paediatric Anaesthesia, KK Women's and Children's Hospital, Singapore; Department of Anaesthesiology, All India Institute of Medical Sciences, New Delhi, India
Click here for correspondence address and email
|Date of Submission||05-Apr-2022|
|Date of Acceptance||16-May-2022|
|Date of Web Publication||27-Jun-2022|
Background: Obstructive sleep apnoea (OSA) in children is associated with the development of perioperative adverse respiratory events. The aim of our study was to find out the incidence of perioperative adverse respiratory events, to identify the risk predictors and to determine the appropriate anaesthetic agents in children with OSA. Methods: After obtaining approval from the the Institutional Review Board, 189 children with OSA who had undergone adenotonsillectomy between 2004 and 2009 were selected from a retrospective review of case files. Variables recorded included demographic data, coexistent medical illnesses, anaesthetic techniques (including induction agents and muscle relaxants) and perioperative analgesia. The severity of OSA was determined based on polysomnographic criteria. Adverse events including difficult airway, desaturation due to bronchospasm or laryngospasm, postoperative desaturation and unplanned intensive care unit (ICU) admission occurring up to 24 h postoperatively were recorded. Results: The incidence of perioperative respiratory adverse events was 19.6%. Severe OSA (odds ratio [OR] 5.8; 95% confidence interval [CI] 1.8–18.53; P = 0.003) and moderate OSA (OR 3.9; 95% CI 1.1–13.1; P = 0.029) were independent risk factors associated with complications. There was no correlation between the intraoperative anaesthetic techniques or use of perioperative opioid analgesics and the perioperative adverse respiratory events. Conclusions: Preoperative diagnosis of OSA using polysomnography has been shown to identify children who are at increased risk of perioperative adverse respiratory events in children. In our review, 19.6% of children with OSA were at risk. The use of a severity index may better identify children at higher risk.
Keywords: Obstructive sleep apnoea in children, perioperative adverse respiratory events, risk factors
|How to cite this URL:|
Thampi S, Chong SY, Pawar DK. Perioperative adverse respiratory events in children with obstructive sleep apnoea. Airway [Epub ahead of print] [cited 2022 Sep 28]. Available from: https://www.arwy.org/preprintarticle.asp?id=348366
| Background|| |
Obstructive sleep apnoea (OSA) is a syndrome characterised by periodic partial or complete obstruction of the upper airways during sleep and can affect up to 3% of the paediatric population., These children are at a higher risk of developing perioperative adverse respiratory events (21% as compared to 1.3% in children without OSA)., Studies determining the risk factors for developing perioperative adverse events in OSA patients have been carried out in adults. There is a paucity of such literature on children. Most studies on OSA have focused mainly on patient characteristics and their comorbidities.
This retrospective review was planned to assess the incidence of perioperative adverse respiratory events, to identify the associated factors and to evaluate the potential association of anaesthetic agents and techniques with the complication rates in children with OSA.
| Methods|| |
Institutional Review Board approval with waiver of informed consent (CIRB Ref: 2010/020/D) was obtained to undertake this review at KK Women's and Children's Hospital, Singapore. Children below the age of 18 years with polysomnogram (PSG) confirmed diagnosis of OSA who underwent adenotonsillectomy between 2004 and 2009 were included in the review. Any child who underwent adenoidectomy or tonsillectomy alone or in combination with uvulopalatoplasty was excluded from the study. Our hospital criteria for diagnosing OSA using PSG are described in the supplemental material. OSA was graded as mild, moderate or severe [Appendix 1].
We reviewed the anaesthetic and perioperative records of these children. Their demographics (age, gender, weight, height and race), coexistent medical illnesses, previous airway manipulation, preoperative use of continuous positive airway pressure (CPAP), type and duration of anaesthesia, and intraoperative anaesthetic techniques including induction agents, muscle relaxants, reversal agents and perioperative analgesics and the use of postoperative lateral position for recovery were recorded. Obesity was defined as a body mass index >95th percentile for age and gender.
We recorded all events of the difficult airway (difficult mask ventilation, difficult intubation or both), intraoperative bronchospasm or laryngospasm causing desaturation. All postoperative desaturation requiring supplemental oxygen and unplanned intensive care unit (ICU) admissions for 24 h after adenotonsillectomy were also recorded. Desaturation was defined as oxygen saturation reading <95% requiring oxygen supplementation. All children had their oxygen saturations monitored using pulse oximetry for the first 24 h postoperatively. Airway manoeuvres such as jaw thrust, airway adjuncts or lateral position were used as indicated to overcome the anticipated problems during difficult mask ventilation.
Data were analysed using SPSS Statistics for Windows, Version 16.0 (SPSS Inc., Chicago, IL, USA).
Continuous variables were analysed using the independent 't'-test and categorical variables using the Chi-square test. Logistic regression was used to determine the association between the risk factors and adverse events. Each of the associated risk factors for perioperative adverse respiratory events was then subjected to univariate analysis to determine a possible association between them and the occurrence of perioperative adverse respiratory events. The odds ratio and 95% confidence interval were computed from logistic regression. Variables that were found to be significantly associated with dependent outcome measures in univariate analysis were then entered into multiple logistic regression models to find out any independent risk factors. P <0.05 was taken to denote statistical significance.
| Results|| |
A total of 189 children identified as having OSA by preoperative polysomnography and who underwent adenotonsillectomy were included in the study. They were further classified as mild (n = 68), moderate (n = 52) and severe (n = 69) OSA based on their sleep study. Children were aged between 4.3 and 12.1 years with a preponderance of male patients (75.1%). Twenty-one of them belonged to the American Society of Anesthesiologists Physical Status (ASA-PS) III or IV. The duration of adenotonsillectomy ranged from 30 min to 90 min. The ethnicity was 53% Chinese, 29% Malay, 10.6% Indians and 7.4% others. Twenty-six of these children had asthma, four children had noncyanotic heart disease and eight were dependent on CPAP. Of these children, 115 were obese. Eleven children had previous airway manipulation and seven had some form of craniofacial anomaly. Demographics and adverse events of the study sample are shown in [Table 1] and [Table 2].
|Table 2: Factors associated with perioperative adverse respiratory events (univariate analysis)|
Click here to view
Out of the 189 patients with OSA, 37 patients had a perioperative adverse respiratory event (19.6%). Out of the 37 patients who developed adverse events, desaturation occurred in six patients intraoperatively and 28 patients in the postanaesthesia care unit or in the ward at various points of time overnight. None of the patients with adverse events had a preexisting upper respiratory tract infection. There were three cases of intraoperative bronchospasm including one in a patient with preexisting asthma (an obese 11 years old). There was a case of difficult mask ventilation but no case of difficult intubation. There were two children with intraoperative laryngospasm.
Variables included in the univariate risk analysis are presented in [Table 2]. Thirty-five per cent of children with severe OSA and 18% with moderate OSA had an adverse event compared to 9% in the mild OSA group. Univariate analysis of the risk factors showed that severe OSA (odds ratio OR, 7.5; 95% confidence interval [CI] 2.4–23.1), moderate OSA (OR, 4; 95% CI 1.1–13.6) and ASA III or IV (OR, 2.95; 95% CI 1.12–7.76) were significantly associated with adverse events (P < 0.05). Multivariate analysis of the factors found to be significant revealed that severe OSA and moderate OSA were independently associated with adverse events [Table 3].
|Table 3: Independent risk factors for perioperative adverse respiratory events (multivariate logistic regression analysis)|
Click here to view
Results showed that 31.7% of children who received an inhalational induction experienced an adverse event. The frequency of adverse events was 19.7% in children who received an intravenous induction. Muscle relaxants were used to facilitate intubation and surgery in 151 cases [Table 2]. Administration of muscle relaxants was not associated with an increased risk of adverse events.
| Discussion|| |
This review of children with identified OSA undergoing adenotonsillectomy under our care was aimed at identifying preoperative, intraoperative and anaesthetic interventions which could be associated with the occurrence of adverse events. As clinical signs and reported symptoms have been shown to be ambiguous in the diagnosis of OSA, we chose to study only the outcome of children who had preoperative polysomnography and had been diagnosed as having OSA. Our study revealed that the incidence of perioperative adverse respiratory events in this group of patients to be 19.6%. This is similar to the results of another study in children with OSA. It is higher than the incidence of respiratory complications (1.3%) previously described in children without a preoperative diagnosis of OSA.
In our study, we did not identify any relationship between the occurrence of adverse events and age, gender, race, body mass index and coexistent medical illnesses. Several studies have previously reported that the incidence of adverse events is higher in a younger age group <2 years.,,, As none of the patients whose records we reviewed was <2 years of age, we were unable to validate a correlation between young age and adverse events.
The severity of OSA is an important predictor of increased risk of adverse events. We observed that irrespective of the anaesthetic technique or the drugs used, severe OSA is associated with more perioperative adverse respiratory events. Paediatric OSA promotes upper airway collapse during inspiration due to both anatomic and dynamic factors. The pressure gradient for this airway collapse between sleep and neuromuscular paralysis is reported to be half that measured in normal children, suggesting that children with OSA show altered neuromuscular control of upper airway patency. We hypothesise that with increasing severity of OSA, there is a higher chance of airway obstruction under anaesthesia. This highlights the importance of determining the severity of OSA in children preoperatively. The diagnostic criteria used for OSA and the determination of its severity vary amongst institutions. Our review suggests that PSG values can be used to predict adverse clinical outcomes in OSA children. Studies by Wilson et al. and Jaryszak et al. have looked at PSG variables in paediatric adenotonsillectomy and concluded that worse PSG scores were associated with more adverse events., Both Rosen et al. and Ye et al. observed that a worse preoperative PSG score predicted higher postoperative respiratory complications., We evaluated the severity of OSA in our subjects on the basis of PSG values. Outcomes from previous studies corroborate that the use of PSG scores for evaluating OSA is reasonable.
ASA-PS status of III or IV also emerged as a risk factor after univariate analysis in our study. This could have been because sicker the patient profile, higher the chances of adverse events. However, it ceased to be significant after performing multivariate analysis.
Previous studies of OSA in children have focused on the relationships between patient characteristics and adverse events. We chose to evaluate whether there was an association between anaesthetic interventions/anaesthetic agents and adverse events. Even though there was a higher incidence of adverse events among those who had an inhalational induction (31.7%) versus an intravenous induction (19.7%), we failed to demonstrate a significant association between the induction technique and perioperative respiratory adverse events. Administration of muscle relaxants was not associated with an increased risk of adverse events.
Waters et al. and Brown et al. demonstrated that children with severe OSA display heightened respiratory sensitivity to intraoperative fentanyl manifesting as apnoea and increased sensitivity to postoperative morphine. Studies also point to the higher incidence of desaturation with morphine compared with dexmedetomidine and more desaturations with pethidine and nalbuphine use in OSA children with a higher PSG score. We compared the perioperative use of long-acting opioids, short-acting opioids or use of nonopioids and did not find any significant difference in the incidence of perioperative adverse respiratory events. This is in agreement with the conclusions drawn by Sanders et al.
It is common practice in our institution to allow paediatric patients to recover in the lateral position as it is believed to relieve airway obstruction better than recovery in the supine position. Lateral positioning has been shown to increase the size of the airway by both magnetic resonance imaging and endoscopy. A lateral recovery position was used in the majority of children studied with only 40 children being allowed to recover in the supine position. We observed that either supine or lateral positioning of the child postoperatively had no correlation with the occurrence of adverse events in our study.
In summary, we evaluated the anaesthetic care and perioperative adverse events in 189 children with OSA who were diagnosed by polysomnography before undergoing adenotonsillectomy. Because this was a retrospective review, we found gaps in data collection impairing our ability to perform a more detailed evaluation of subjects in our study. The effect of premedication on adverse events could not be included due to nonavailability of sufficient data in anaesthetic charts. We could not compare the effects of using different inhalation agents as most of the patients received only sevoflurane. We could not capture data on deep versus awake extubation as the documentation was lacking in most of the anaesthetic charts. Further prospective studies with more detailed information of perioperative anaesthetic management (premedication drugs used, method of anaesthetic induction, position for extubation and deep versus awake extubation) would add a lot of information to our current knowledge of paediatric OSA.
We believe that polysomnography should be performed preoperatively in clinically diagnosed OSA as it helps to predict the occurrence of perioperative adverse respiratory events. This will help in counselling the patient and family, giving them realistic hospital expectations, planning anaesthetic management strategies and providing focused postoperative care.
| Conclusions|| |
Preoperative diagnosis of OSA using polysomnography has been shown to identify children who are at increased risk of perioperative adverse respiratory events. Where OSA has been identified, we found the risk of adverse respiratory events to be 19.6%. The use of a severity index may better identify children who are in a higher-risk category than others. Coexisting medical illnesses, anaesthetic technique and perioperative opioids were not independent risk factors in our study for the occurrence of perioperative adverse events in children with OSA. We believe that information obtained from the preoperative conduct of polysomnography may possibly be used for risk stratification, parental counselling, planning of anaesthetic management and postoperative disposition in these children.
We acknowledge with thanks continuous help provided by Siti Nurdiyanah (Research Assistant, Department of Paediatric Anaesthesia) and Valli from Medical Records Section, KK Women's and Children's Hospital throughout the study.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
Appendix 1: Polysomnography (PSG)
Sleep studies were performed and reported in our hospital in the Paediatric Sleep laboratory in accordance with the American Academy of Sleep Medicine guidelines. Children were admitted and observed in the sleep laboratory overnight. A minimum of 8 h was recorded.
The PSG montage included
- Electroencephalogram (EEG) monitoring
- Electrooculogram (EOG) monitoring
- Electromyogram (EMG) monitoring
- Airflow parameters – thermocouple, pressure transducer and transcutaneous CO2 analyser
- Respiratory effort parameters – thoracic and abdominal inductance belts
- Oxygen saturation
- Body position
- Electrocardiogram (ECG) monitoring.
Respiratory variables, sleep architecture and arousal scoring were investigated. Children were graded as mild, moderate or severe OSA based on the OAHI. OAHI (Obstructive apnoea–hypopnea index) is the average number of apnoea and hypopnoeas per hour of sleep. The airflow through the nose was monitored and any apnoea (defined as ≥90% reduction in airflow associated with continued inspiratory effort throughout the entire period of absent airflow) or hypopnea (defined as a ≥30% reduction in airflow associated with either arousal or ≥3% desaturation) were recorded.
| References|| |
Gross JB, Bachenberg KL, Benumof JL, Caplan RA, Connis RT, Coté CJ, et al.
Practice guidelines for the perioperative management of patients with obstructive sleep apnea: A report by the American Society of Anesthesiologists Task Force on perioperative management of patients with obstructive sleep apnea. Anesthesiology 2006;104:1081-93.
Lumeng JC, Chervin RD. Epidemiology of pediatric obstructive sleep apnea. Proc Am Thorac Soc 2008;5:242-52.
Wilson K, Lakheeram I, Morielli A, Brouillette R, Brown K. Can assessment for obstructive sleep apnea help predict postadenotonsillectomy respiratory complications? Anesthesiology 2002;96:313-22.
Richmond KH, Wetmore RF, Baranak CC. Postoperative complications following tonsillectomy and adenoidectomy – Who is at risk? Int J Pediatr Otorhinolaryngol 1987;13:117-24.
Certal V, Catumbela E, Winck JC, Azevedo I, Teixeira-Pinto A, Costa-Pereira A. Clinical assessment of pediatric obstructive sleep apnea: A systematic review and meta-analysis. Laryngoscope 2012;122:2105-14.
Rosen GM, Muckle RP, Mahowald MW, Goding GS, Ullevig C. Postoperative respiratory compromise in children with obstructive sleep apnea syndrome: Can it be anticipated? Pediatrics 1994;93:784-8.
Hill CA, Litvak A, Canapari C, Cummings B, Collins C, Keamy DG, et al.
A pilot study to identify pre- and peri-operative risk factors for airway complications following adenotonsillectomy for treatment of severe pediatric OSA. Int J Pediatr Otorhinolaryngol 2011;75:1385-90.
Ye J, Liu H, Zhang G, Huang Z, Huang P, Li Y. Postoperative respiratory complications of adenotonsillectomy for obstructive sleep apnea syndrome in older children: Prevalence, risk factors, and impact on clinical outcome. J Otolaryngol Head Neck Surg 2009;38:49-58.
Katz SL, Monsour A, Barrowman N, Hoey L, Bromwich M, Momoli F, et al.
Predictors of postoperative respiratory complications in children undergoing adenotonsillectomy. J Clin Sleep Med 2020;16:41-8.
Katz ES, D'Ambrosio CM. Pathophysiology of pediatric obstructive sleep apnea. Proc Am Thorac Soc 2008;5:253-62.
Jaryszak EM, Shah RK, Vanison CC, Lander L, Choi SS. Polysomnographic variables predictive of adverse respiratory events after pediatric adenotonsillectomy. Arch Otolaryngol Head Neck Surg 2011;137:15-8.
Waters KA, McBrien F, Stewart P, Hinder M, Wharton S. Effects of OSA, inhalational anesthesia, and fentanyl on the airway and ventilation of children. J Appl Physiol (1985) 2002;92:1987-94.
Brown KA, Laferrière A, Moss IR. Recurrent hypoxemia in young children with obstructive sleep apnea is associated with reduced opioid requirement for analgesia. Anesthesiology 2004;100:806-10.
Zhuang PJ, Wang X, Zhang XF, Zhou ZJ, Wang Q. Postoperative respiratory and analgesic effects of dexmedetomidine or morphine for adenotonsillectomy in children with obstructive sleep apnoea. Anaesthesia 2011;66:989-93.
Habre W, McLeod B. Analgesic and respiratory effect of nalbuphine and pethidine for adenotonsillectomy in children with obstructive sleep disorder. Anaesthesia 1997;52:1101-6.
Sanders JC, King MA, Mitchell RB, Kelly JP. Perioperative complications of adenotonsillectomy in children with obstructive sleep apnea syndrome. Anesth Analg 2006;103:1115-21.
Litman RS, Wake N, Chan LM, McDonough JM, Sin S, Mahboubi S, et al.
Effect of lateral positioning on upper airway size and morphology in sedated children. Anesthesiology 2005;103:484-8.
Iber C, Ancoli-Israel S, Chesson Jr., AL, Quan SF. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. Version 2.0. Am Acad Sleep Med 2007:1-32.
Department of Anaesthesia, National University Hospital, 5 Lower Kent Ridge Road
Source of Support: None, Conflict of Interest: None
[Table 1], [Table 2], [Table 3]
| Article Access Statistics|
| Viewed||255 |
| PDF Downloaded||7 |