Preoperative continuous positive airway pressure use improves the hospital experience of children undergoing adenotonsillectomy for obstructive sleep apnea

INTRODUCTION

In children the most common cause of obstructive sleep apnea (OSA) in children is adenotonsillar hypertrophy, and as such the primary treatment recommendation is adenotonsillectomy.^1-6 However, children having adenotonsillectomy for OSA are up to five times more likely to have respiratory complications post operatively compared to children without OSA.⁷ Major post-operative respiratory complications in children with OSA undergoing adenotonsillectomy include laryngospasm, bronchospasm, pulmonary edema, pneumonia and requirement for re-intubation.⁸ The risk for these perioperative complications increases with increasing severity of OSA.⁸

De et al demonstrated significant residual sleep disordered breathing and altered sleep architecture on the first night following adenotonsillectomy, with similar values for AHI, oxygen nadir, arousal index and sleep efficiency to the preoperative PSG.

Other modalities for treatment of OSA include intranasal steroids and/or oral montelukast for mild OSA. More severe disease may require therapy with continuous positive airway pressure (CPAP) or additional surgery such as lingual tonsillectomy.⁶

To date, the primary recommendations for use of CPAP in children are for persistent disease following adenotonsillectomy, with some reports of the post-operative use of CPAP or BiLevel to manage medically complex children with post-operative respiratory complications.^1,10

We found no previous studies evaluating the use of CPAP prior to adenotonsillectomy in children. This retrospective study evaluated how a recommendation for pre-operative use of CPAP impacted the peri-operative course of children undergoing adenotonsillectomy for OSA, including length of wait for surgery, post-operative use of respiratory support, admission to PICU, postoperative oxygen saturation, length of stay in hospital, and readmission to hospital.

METHODS

A retrospective chart review was undertaken at an Australian tertiary pediatric hospital. Inclusion criteria were children diagnosed with OSA on PSG up to two years prior to their adenotonsillectomy, in the period between January 2011 and December 2017. The study was approved by The Sydney Children’s Hospital Network Human Research Ethics Committee (LNR/18/SCHN/401). The presence of comorbidities was documented and categorised to include trisomy 21, craniofacial disorders, neuromuscular disorders, cardiac and respiratory disease, FTT, obesity, and upper-airway abnormalities.

Severity of OSA was defined as mild (AHI=1-5), moderate (AHI=5-10) and severe (AHI>10).¹¹ Patients were categorized into three groups; not prescribed CPAP (N-CPAP), prescribed and non-compliant with CPAP (NC-CPAP) and compliant with a prescription for CPAP (C-CPAP). For analysis of outcomes, the N-CPAP group was considered the control group. The recommendation to initiate CPAP therapy was made by the treating clinician, based on their evaluation of the PSG and clinical data. Compliance with CPAP therapy was based on parental report of regular use of CPAP in the two months preceding surgery.

Length of wait for surgery was defined as the period between the clinic date in which the decision was made to proceed with adenotonsillectomy and the date when surgery was performed. The recovery period was defined as the first 6 hours and the post-operative period as the period in hospital after the first 6 hours. Respiratory support requirements were divided into four categories; low flow oxygen, high flow oxygen by nasal prongs, non-invasive ventilation (NIV) which included CPAP and BiPAP, and intubation.

Outcomes

Primary outcomes were admission to pediatric intensive care unit (PICU) and requirement for respiratory support in the recovery and postoperative period(s). Secondary outcomes included length of wait for surgery, minimum saturations noted in hospital, length of stay in hospital, and readmission to hospital. Subgroup analyses of both primary and secondary outcomes was conducted for the group of patients who were not prescribed CPAP, and on the patients diagnosed with severe OSA, regardless of whether they were prescribed CPAP pre-operatively.

Statistical analysis

Statistical analysis was performed using SAS 9.4 Descriptive statistics were presented as mean for continuous variables and frequencies and percentages for categorical variables. Patient characteristics and study outcomes were compared between groups using Kruskal-Wallis tests for continuous variables or linear models for log-transformed continuous variables, and chi-square tests or logistic regression for categorical variables. Statistical significance was defined as a p value of <0.05.

RESULTS

Baseline characteristics

Of 216 eligible patients, 162 were included in the study group. 50 patients were excluded due to missing data, 2 were excluded for use of CPAP prior to first available PSG and 2 due to having a tracheostomy. Of the 162 studied, 33% were female (n=54) and mean age was 5.5 years (10 months to 16 years). OSA was mild in 59, moderate in 37 and severe in 66 (Table 1).

CPAP was prescribed for 55 patients, 30 were C-CPAP and 25 were NC-CPAP. Mean patient age differed between groups with N-CPAP 6yo, NC-CPAP 3.9 yo and C-CPAP 5 yo (p=0.02). Markers of OSA severity were worse in the groups prescribed CPAP. Mean total AHI was higher; N-CPAP AHI 6.8, NC-CPAP 22.3 and C-CPAP 23.3 (p<0.0001). Oxygen nadir was lower; N-CPAP 87%, NC- CPAP 75% and C-CPAP 80% (p<0.0001). Median prescribed CPAP pressure was 6 cm H₂0 (SD±1.4); with a range of 5-11 cmH₂0 and 5-9 cm H₂0 for the NC-CPAP and C-CPAP groups, respectively (p=0.45). Amongst the comorbidities, FTT was more frequent in those prescribed CPAP (p=0.048) while non-specified comorbidities were more frequent in the N-CPAP group 56% compared to NC-CPAP 18% and C-CPAP 26% (p=0.03). Baseline characteristics are outlined in Table 1.

1% of total number in OSA subcategory, 2% of total with comorbidities within subcategory, CPAP – continuous positive airway pressure, BMI – body mass index, AHI – apnoea hypopnoea index, OAHI – obstructive apnoea hypopnoea index, PSG – polysomnography, CO₂ – carbon dioxide, FTT – failure to thrive

PICU – Paediatric intensive care unit, CPAP – continuous positive airway pressure, CI – confidence interval, NIV – non-invasive ventilation

Primary outcomes

Comparisons were made against the group not prescribed CPAP (N-CPAP). Of the 162 patients, 20 (12.3%) were admitted to PICU; with increased risk for PICU admission if prescribed CPAP, and greatest in NC-CPAP (OR 12, CI 3-44) compared to C-CPAP (OR 9, CI 2-33) p=0.0003. During the recovery period, low flow oxygen use was equivalent amongst the 3 groups, but those compliant with CPAP were more likely to use NIV compared to N-CPAP and NC-CPAP groups (p<0.0001). During the subsequent postoperative period, the C-CPAP group were less likely to have low flow oxygen NC-CPAP 36% and N-CPAP 30% p=0.003. Those prescribed CPAP were more likely to use NIV, NC-CPAP 24% and C-CPAP 67% p=0.0001. There was no difference in risk of intubation between the groups during the recovery and postoperative periods. In addition, logistic regression analysis demonstrated the risk of PICU admission was not confounded by underlying severity of OSA. Primary outcomes are shown in Table 2.

Secondary outcomes

For children prescribed CPAP, wait time to surgery was lower; N-CPAP (29.6 weeks) NC-CPAP 15.8 weeks (47% reduction, p=0.0008) and C-CPAP 21.7 weeks. Length of hospital stay was greater in those prescribed CPAP, N-CPAP 0.9 days, NC-CPAP greatest at 2.1 days (59% increase p=0.001), C-CPAP with LOS 1.2 days (34% increase p=0.03). Minimum oxygen saturation in the recovery period was lower in those prescribed CPAP; N-CPAP 96.7%, NC-CPAP 95.6% and C-CPAP 92.8% (p=0.0002). No differences were seen for minimum oxygen saturations in the postoperative period, or readmission rates (Table 3).

¹Ratio of linear regression of log length of stay, CPAP – continuous positive airway pressure, Av – average, CI – confidence interval

¹Ratio of linear regression of log length of stay, CPAP – continuous positive airway pressure, PICU – paediatric intensive care unit, CI – confidence interval, NI – non-invasive ventilation, NS – not significant, Av – average

CPAP – Continuous positive airway pressure, OSA – obstructive sleep apnoea, PICU – paediatric intensive care unit, NIV – non-invasive ventilation, Av – average, CI – confidence intervals

Subgroup analyses

Subgroup analysis for only those patients diagnosed with severe OSA (AHI>10) demonstrated a greater use of NIV support in those prescribed CPAP; NC-CPAP 29% and C-CPAP 64% (p<0.0001) compared to those not prescribed CPAP (Table 4). Amongst those not prescribed CPAP, length of wait for surgery was reduced in those with severe OSA at 21.3 weeks (35% reduction, p=0.04) compared to mild OSA (30.4 weeks) and moderate OSA (34.6 weeks, p=0.56) patients (Table 5).

DISCUSSION

Our long-established experience with use of CPAP in children has led to processes for pre-operative implementation of CPAP in children with OSA.¹² In view of the acknowledged wait times for surgery, the use of CPAP offers an immediate intervention to treat children with more severe disease, and this practice permitted the current retrospective evaluation of this system of care. We found that compliance with pre-operative CPAP led to improved use of CPAP in the immediate postoperative period, along with reduced use of oxygen and shorter post-operative stay compared to those prescribed, but non-compliant with, CPAP.

Preoperative risk assessment prior to adenotonsillectomy plays an important role in elective, postoperative admission to PICU. Recognized risk factors for post-operative respiratory complications following adenotonsillectomy for OSA include age <3 yo, severe OSA, failure to thrive, obesity, cardiac complications of OSA, trisomy 21, prematurity, craniofacial abnormalities, neuromuscular disorders, chronic lung disease and sickle cell disease.^1,13,14 Even using these criteria to define risk groups, many of those admitted to PICU require only routine care and do not require high dependency or intensive care management.¹⁵

In the current study group, all admissions to PICU were planned, but compliance with CPAP therapy resulted in fewer PICU admissions. We believe this is most likely because their preoperative use of CPAP influenced their perceived postoperative risks when they were assessed prior to surgery. Children prescribed CPAP were younger than those not prescribed therapy, consistent with the finding that younger children have more severe disease, however, the non-compliant children were also the youngest.^16,17

Compliance with preoperative CPAP therapy was associated with an overall reduction in hospital length of stay by 1 day, compared to those who were non-compliant. Though cost was not specifically assessed in the study, we suggest that compliance with CPAP therapy would positively impact on resource allocation and cost effectiveness for the hospital system. Further studies looking into resource requirement for establishment of CPAP and comparing with cost of PICU admission and average cost of general hospital admission should be considered.

Saur et al reviewed studies of populations with high-risk for post-operative complications, highlighting that recognition of these children’s’ increased post-operative risk means that additional care is taken to identify and manage potential respiratory complications should they arise. The authors also highlighted that multiple assessments, including sleep studies, can create delays treatment intervention. Wait times can be long for both sleep medicine and ear, nose and throat (ENT) surgical services. Our rationale for prescribing preoperative CPAP includes the opportunity for intervention earlier than the surgery would allow which, even in those with severe OSA who were unable to use CPAP, was >14 weeks.

The overall low rate of respiratory complications in our study group, despite the pre-operative identification of their risk factors, is consistent with recent studies.^14,18 Nonetheless, the only major respiratory complication, re-intubation, was seen in one case of each of our groups (N-CPAP, NC-CPAP and C-CPAP), the only difference amongst the groups for post-operative respiratory complications, was higher use of NIV in children who were previously prescribed CPAP. We postulate that the reason postoperative CPAP use was highest in children who were compliant with therapy pre-operatively, was its ease of use in children previously acclimatized to the therapy since many post-operative instructions in the medical records recommended the use of the patients’ own CPAP postoperatively, or directly after extubation in theatres. Consistent with other studies of children undergoing sleep studies prior to adenotonsillectomy, the majority of children in this study had their surgery at our hospital because they were identified as high risk for respiratory compromise after adenotonsillectomy for OSA.¹⁹ In such a baseline high-risk population, another study found that severity of OSA on PSG was the distinguishing factor for those with significant respiratory complications.¹⁸ Children have a two-fold increased incidence of fatal respiratory events following adenotonsillectomy, with strong evidence that severity of OSA is an important determinant of postoperative risk.¹⁹ Thus, the lack of difference in other postoperative complications or adverse outcomes may be a positive result of the pre-operative CPAP usage, although our study numbers are not sufficient to consider the outcome proven.

In keeping with published literature, compliance with CPAP did not improve the minimum oxygen saturations, nor prevent requirement for respiratory support in the postoperative period. Postoperative respiratory complications occur in up to 27% of patients with OSA, including desaturations and hypopnea, and De et al demonstrated no difference in oxygen nadir on PSG the first night after adenotonsillectomy compared to preoperative PSG.²⁰ One of the few mentions of pre-operative CPAP use in children recommends that those on CPAP preoperatively should continue on CPAP postoperatively to reduce risk of respiratory complications.²⁰

Future considerations

CPAP therapy improves gas exchange, cardiovascular and metabolic risks, neurocognitive effects and quality of life in children with OSA.^21-24 However, current recommendations for the use of CPAP in children with OSA secondary to adenotonsillar hypertrophy focus on its use in the presence of residual OSA despite adenotonsillectomy.^1,3 Adenotonsillar hypertrophy is the leading cause of OSA in children, and adenotonsillectomy is the primary treatment modality, however the prolonged length of wait for surgery especially in our public health system, suggests that CPAP therapy may have a role in managing severe OSA while children are awaiting surgery. Additional studies will be required to fully define the role this therapy in the pre-operative setting, particularly in children who have high risk for persisting disease.

Limitations

As a retrospective study we relied on information in the medical records of the patients at the time of their admission for adenotonsillectomy, resulting in 25% loss of numbers through inadequate information. Nonetheless, we were able to demonstrate that pre-operative compliance with a recommendation for CPAP therapy resulted in higher use of CPAP support in the recovery and post-operative periods, and a 1-day reduction in the duration of hospital stay. The authors acknowledge the selection bias within the included patient population, and that the patient groups may not be reflective of the general population. Additionally, a significant limitation, due to the retrospective nature of the study, was the lack of objective data regarding compliance with CPAP use. Compliance is generally defined as use of CPAP for a minimum of 4 hours per night, with current technology, electronic data downloads are now available for most devices. Of the study population, only 16 % had compliance downloads available; for the remainder of the study population compliance was based on documented verbal reports from chart reviews, which the authors acknowledge creates a possibility of bias.

CONCLUSION

In children undergoing adenotonsillectomy for OSA diagnosed on PSG, preoperative compliance with CPAP therapy reduced the risk of admission to PICU, and resulted in a shorter length of hospital admission, with less requirement for oxygen support in the postoperative period.

ACKNOWLEDGEMENTS

The authors would like to acknowledge Ms. Elizabeth Barnes for assistance with statistical analysis for the study.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. Marcus C, Brooks L, Davidson Ward S, Draper K, Gozal D, Halbower A, et al. Diagnosis and Management of Childhood Obstructive Sleep Apnea Syndrome. Pediatrics. 2012;130(3):714-55.
2. 2. Horne R, Yang J, Walter L, Richardson H, O’Driscoll D, Foster A, et al. Elevated blood pressure during sleep and wake in children with sleep-disordered breathing. Paediatrics. 2011;128(1):85-92.
3. 3. Dehlink E, Tan H. Update on paediatric obstructive sleep apnoea. J Thorac Dis. 2016;8(2):224-35.
4. 4. Alsubie H, BaHammam A. Obstructive Sleep Apnoea: Children are not little Adults. Paediatric Resp Rev. 2017;21:72-9.
5. 5. Ehsan Z, Ishman S. Pediatric Obstructive Sleep Apnea. Otolaryngol Clin N Am. 2016;49:1449-64.
6. 6. Brockbank JC. Update on pathophysiology and treatment of childhood obstructive sleep apnea syndrome. Paediatric Resp Rev. 2017;24:21-3.
7. 7. De Luca Canto G, Pacheco-Pereira C, Aydinoz S, Bhattacharjee R, Tan H, Kheirandish-Gozal L, et al. Adenotonsillectomy Complications: A Meta-analysis. Pediatrics. 2015;136(4):702-18.
8. 8. Kang K, Chang I, Tseng C, Weng W, Hsiao T, Lee P, et al. Impacts of disease severity on postoperative complications in children with sleep-disordered breathing. Laryngoscope. 2017;127:2646-52.
9. 9. De A, Waltuch T, Gonik N, Nguyen-Famulare N, Muzumdar H, Bent J, et al. Sleep and breathing the first night after adenotonsillectomy in obese children with obstructive sleep apnea. J Clin Sleep Med. 2017;13(6):805-11.
10. 10. Buzi A, Coblens OM, MacGillivray M, Elden L. The Use of Noninvasive Positive Pressure Ventilation following Pediatric Tonsillectomy. Otolaryngol–Head Neck Surg. 2017;157(2):297-301.
11. 11. Ehsan Z, Ishman SL. Pediatric Obstructive Sleep Apnea. Otolaryngol Clin N Am. 2016;49:1449-64.
12. 12. Waters KA, Everett FM, Bruderer JW, Sullivan CE. Obstructive sleep apnea: the use of nasal CPAP in 80 children. Am J Respir Crit Care Med. 1995;152(2):780-5.
13. 13. Patino M, Sadhasivam S, Mahmoud M. Obstructive sleep apnoea in children: perioperative considerations. Br J Anaesth. 2013;111(51):83-95.
14. 14. Saur J, Brietzke S. Polysomnography results versus clinical factors to predict post-operative respiratory complications following pediatric adenotonsillectomy. Int J Pediatric Otorhinolaryngol. 2017;98:136-42.
15. 15. Levi E, Alvo A, Anderson B, Mahadevan M. Postoperative admissions to paediatric intensive care after tonsillectomy. Sage Publications. 2020;8:1-5.
16. 16. Bains A, Abraham E, Hsieh A, Rubin B, Levi J, Cohen M. Characteristics and frequency of children with severe obstructive sleep apnea. Otolaryngol Head Neck Surg. 2020;163(5);1055-60.
17. 17. Selvadurai S, Voutsas G, Propst E, Wolter N, Narang I. Obstructive sleep apnea in children aged 3 years and younger: Rate and risk factors. Paediatr Child Health. 2019;25(7):432-8.
18. 18. Billings KR, Somani SN, Lavin J, Bhushan B. Polysomnography variables associated with postoperative respiratory issues in children <3 Years of age undergoing adenotonsillectomy for obstructive sleep apnea. Int J Pediatr Otorhinolaryngol. 2020;137:110215.
19. 19. Brown K. Outcome, risk, and error and the child with obstructive sleep apnea. Pediatric Anesthesia. 2011;21:771-80.
20. 20. Patino M, McAuliffe J, Francis L, Mohamed M. Obstructive Sleep Apnea syndrome: Anesthetic Concerns. Int Anesthesiol Clin. 2012;50(4):41-53.
21. 21. Kaditis A, Alvarez M, Boudewyns A, Alexopoulos E, Ersu R, Joosten K, et al. Obstructive Sleep Disordered Breathing in 2- to 18-year-old children: diagnosis and management. Eur Respir J. 2016;47:69-94.
22. 22. Marcus C, Radcliffe J, Konstantinopoulou S, Beck S, Cornaglia MA, Traylor J, et al. Effects of positive airway pressure therapy on neurobehavioral outcomes in children with obstructive sleep apnea. Am J Respir Crit Care Med. 2012;185(9):998-1003.
23. 23. Beebe D, Byars K. Adolescents with obstructive sleep apnea adhere poorly to positive airway pressure (PAP), but PAP users show improved attention and school performance. PLoS One. 2011;6(3):e16924.
24. 24. Parmar A, Baker A, Narang A. Positive airway pressure in pediatric obstructive sleep apnea. Paediatric Resp Rev. 2019;31:43-51.