Pulmonary edema post-adenotonsillectomy in children

Etiology/pathophysiology

Post-adenotonsillectomy pulmonary edema may be considered as negative pressure pulmonary edema (NPPE) which has 2 major subclasses. The first is type I-NPPE which results from the forceful expiration against an acutely obstructed upper airway. The second is type II-NPPE which results from the surgical removal of a chronic upper airway obstruction, such as in an OSA patient who underwent adenotonsillectomy. This results in a sudden decrease in intrathoracic pressure precipitating transudation of fluid into the interstitial and alveolar spaces. Reasons for the development of type I-NPPE include accidental occlusion of an endotracheal tube while intubated,20 croup and epiglottitis especially in children,21 and laryngospasm due to post-extubation.22 In addition, administration of muscle relaxants at the beginning of an inhalational induction of anesthesia can induce type I-NPPE due to the premature paralysis of glossal muscles before the diaphragm.23 Generally, the symptoms of PE develop immediately, although sometimes the onset may be considerably delayed for a few hours in the postoperative period. Thus, close post-operative observation must be continued for an extended time in patients with risk factors for pATPE.19

A couple of widely accepted mechanisms have been published to demonstrate the development of pATPE. One mechanism suggests that chronic upper airway obstruction associated with OSA promotes an elevated level of auto positive end-expiratory pressure (auto-PEEP). Type II-NPPE develops upon sudden relief for chronic upper airway obstruction via adenotonsillectomy (sudden decrease in auto-PEEP), creating a negative intrathoracic pressure (-50 to -100 cmH₂O). Consequently, venous return to the heart is increased with subsequent elevation of ventricular end-diastolic volume. Thus, the left ventricular end-diastolic pressure is elevated leading to transudation of fluids in the lung interstitium and alveoli causing pulmonary edema.24-26 On the other hand, type I-NPPE originates from negative intrathoracic pressure developed from forceful respiration against an acutely-obstructed upper airway. Furthermore, obstruction resulting in hypoxia in both types triggers a massive sympathetic discharge causing a blood shift from the systemic to the pulmonary circulation due to negative intrapulmonary pressure.27 Another suggested mechanism states that the chronic mechanical stress developed from respiration against an obstructed upper airway may induce alterations in the alveolar epithelial and pulmonary microvascular membranes, resulting in increased pulmonary capillary permeability, and consequently, protein-rich PE.25,28,29

Epidemiology

Post-operative adenotonsillectomy complications may be classified into respiratory and non-respiratory complications. Respiratory complications include persistent oxygen desaturation, laryngospasms, bronchospasms, and pulmonary edema. Generally, post-operative respiratory complications are approaching 6.5% in children; while children less than 3 years old have a rate of 10%.30 However, pATPE is a rare, life threatening complication that requires immediate intervention.31-33

In a study of 86 post-adenotonsillectomy children with a mean age of 5.3±2.2 years and without craniofacial anomalies or significant cardiopulmonary comorbidities, cases with apnea-hypopnea index (AHI) ≥40 showed significant statistical correlations with post-operative respiratory complications.34 Results of the study revealed 11.6% had respiratory complications, none of which was pulmonary edema. Kon-Tai et al35 conducted a study of 610 children aged <18 years, mean age 7.2 ±3.3 years, with diverse comorbidities and risk factors. All underwent adenotonsillectomy for the treatment of OSA. Although the study reported complications related to oxygen saturation-related between 1.6% to 14.3% with varying severity, only 0.5% developed pneumonia. The Kon-Tai study revealed a low risk of major respiratory complications (laryngospasm, bronchospasm, pulmonary edema, pneumonia) following adenotonsillectomy in this population. Melissa et al30 reported 6.4% as prevalence of post-adenotonsillectomy respiratory complications in a study that involved 2315 patients under 6 years. Of this population, 0.7% developed serious complications including atelectasis, infiltrate, and pulmonary edema. Sofia et al36 documented a 0.45% prevalence of PE among a study cohort of 221 school-aged children. Nuntiger et al37 also conducted a study involving 216 patients with an average age of 6.6 years and high proportions of comorbidities such as craniofacial abnormalities, obesity, and pulmonary hypertension. In this study, a 4.3% prevalence of post-operative PE after adenotonsillectomy was reported. Guinard referred to Type I-NPPE that develops due to laryngospasm after airway surgery as laryngospasm-induced pulmonary edema.27 The Australian Incident Monitoring Study of 4000 incidents of laryngospasm during anesthesia showed that NPPE occurs in up to 4% of all incident reports of laryngospasm.38 The previous high ratios could be attributed to the wide inclusion criteria of the study regarding comorbidities and type of surgical procedure.

Risk factors for development of pATPE

Development of pulmonary edema after adenotonsillectomy is life threatening condition; however, it is uncommon complication. Young age (<3 years) or the presence of some comorbid conditions such as asthma, history of prematurity and craniofacial malformations may increase the risk of developing PE in children undergoing adenotonsillectomy.30

Asthma is one of the risk factors for the development of PE after adenotonsillectomy, as it predisposes pulmonary capillary epithelium to disruption either directly due to high breathing workload, or indirectly by cytotoxic proteins derived from the migration of inflammatory cells. Regarding etiological mechanisms, a pathophysiological cascade due to any of these mechanisms leads to an extravasation of proteins from post-capillary venules into interstitial tissues resulting in the development of PE.39-41

Premature infants with respiratory distress syndrome (RDS) who were ventilated at birth commonly have a condition known as chronic lung disease (CLD) of prematurity. From a histological point of view, Hulsmann R. and Van Anker42 stated that the evolution of CLD can be divided into an early inflammatory phase followed by a subacute and chronic fibroproliferative phase. An influx of inflammatory cells and increased levels of cytokines in alveolar sacs characterize the first phase, while the second phase is characterized by hyperplasia of type II pneumocytes, and interstitial and perialveolar fibrosis. Surviving patients show persistent lung function abnormalities. Therefore CLD is considered a risk factor for pATPE due to abnormalities of alveolar wall.

Alida et al43 reported that 95% of infants with Down syndrome who were referred for PSG had OSA, out of which 75% had severe OSA, which qualified them as candidates for adenotonsillectomy. Bronchial asthma, Down syndrome, and other craniofacial conditions may precipitate high auto-PEEP, which upon sudden relief via adenotonsillectomy result in the development of pATPE.44

Management

As a subtype of pulmonary edema, pATPE is usually self-limited and easily corrected.45 Supportive therapy in these patients begins with close observation in an ICU or step-down unit that has high surveillance. Strategies for correction of pATPE include maintenance of airway via intubation. If the airway is obstructed due to the patient biting down on the endotracheal tube, a dose of succinylcholine (0.1-0.2 mg/kg) may be given to relax the jaw muscles.19 Invasive or non-invasive supplemental oxygen to restore appropriate PEEP via continuous positive airway pressure (CPAP) or bi-level positive airway pressure (BPAP) should be provided. Generally, diuretics may be administered to correct fluid accumulation in lung interstitial spaces and alveoli, unless the patient is hypovolemic. In severe cases of pATPE, mechanical ventilation may be required. It is worth mentioning that non-invasive respiratory support via CPAP or BPAP decreases the work of breathing, improves alveolar gas exchange, reduces left ventricular preload and afterload, increases cardiac output and improves hemodynamics.19 Evidence suggests that noninvasive respiratory support may be an effective strategy to reduce intubation rates,46,47 as well as ICU and hospital lengths of stay46 in postoperative patients.19

Perioperative measures to minimize the probability of pATPE and ICU admission rates.

Generally, adenotonsillectomy is an out-patient procedure; the patient may be discharged on the same day after procedure regardless the surgical technique utilized. However, in some situations it is recommended to hospitalize patient for few hours to a few days based on clinical evaluation and the presence of comorbid conditions.30 Specifically, children at risk for upper airway obstruction or respiratory compromise following tonsil and adenoid surgery such as those under the age of 3, or who have comorbidities like neurological problems, congenital heart disease, prematurity, Down syndrome or craniofacial abnormalities are candidates for post-adenotonsillectomy admission. It has been found that 80% of pulmonary edema occurs within minutes after relief of upper airway obstruction.48 Edema ranges from mild to severe depending on the degree of obstruction and pulmonary capillary injury.45 Cost reduction is a worldwide trend in the health care system. This current work involves a cost/benefit assessment of OSA patients who underwent adenotonsillectomy and were admitted to the ICU as a precaution for the possibility of developing pulmonary edema. Patient variation and waste associated with the delivery of care is a major factor that contributes to the escalation of healthcare costs. In order to minimize ICU admission rates, it is necessary to analyze the root causes leading to ICU admissions. In this context, the optimum management of healthcare resources requires, first of all, the assessment of the cluster of patients with OSA who require in-patient admission. Secondly, it is important to highlight the tools and precautions that can reduce rates of ICU admission even among such a risky group (Table 1).

Preoperative clinical assessment

A pre-operative clinical assessment of patients could dramatically help rule out children with comorbid risk factors, generally decreasing the rate of post-operative ICU admissions. It is important to mention the controversial criteria of the American Academy of Otolaryngology (AAO) and American Academy of Pediatrics (AAP) which recommend overnight monitoring of all OSA patients with AHI ≥10 (by the AAO) and AHI ≥24 (by the AAP) following an adenotonsillectomy. As discussed in the next section, simple monitoring of pulse oximetry could empirically help solve such controversies surrounding the potential development of PE. Specifically, the next few paragraphs will focus on children who do have risk factors for development of pATPE.

Operative/postoperative monitoring tools

With prompt diagnosis and intervention, most pATPE patients can be treated without complications. Since PE develops in a matter of minutes after adenotonsillectomy, simply monitoring the oxygen saturation in the operating room and post-anesthesia care unit using pulse oximetry usually identifies patients at risk and in need of in-patient hospitalization rather than ICU admission. This is because signs and symptoms in most cases are subtle.45 On the other hand, in the absence of pulse oximetry, in-patient admission of patients with OSA could be limited to patients who have risk factors for the development of pATPE. Such patients include children younger than 3 years old, patients with severe OSA, those with comorbid conditions including asthma, obesity, cardiac involvement (right ventricular hypertrophy), Down syndrome, history of prematurity, craniofacial abnormalities, neuromuscular diseases, chronic lung disease, and sickle cell disease.49-55

In conclusion, a combination of pulse oximetry monitoring and clinical evaluation of patients can effectively identify cases in need of in-patient monitoring. Since the management of pATPE ranges from simple oxygen supplementation, application of noninvasive positive airway pressure up to diuretic therapy, advent of inexpensive portable versions of pulse oximeters allow close patient monitoring for pATPE development in sub-ICU level care units, stepping up and down to ICU or normal ward care according to case progression. Primarily, patients with risk factors may be admitted to sub-ICU level care units provided there are no signs of oxygen desaturation and other vital signs. Such simple surveillance could dramatically reduce the cost of healthcare services. Intensive care unit admissions should be limited to cases that require multiple-organ support or advanced respiratory support such as severe PE in which pulmonary venous pressure exceeds 30 mm Hg56 and which requires invasive mechanical ventilation/CPAP/BPAP via trans-laryngeal tracheal or tracheostomy.

Procedural “Surgical-anesthetical” precautions

Several surgical techniques have been developed over the years, but the core principles remain unchanged. The achievement of successful surgery starts with induction of anesthesia, patient positioning for oral intubation, and the surgical dissection technique. Extracapsular techniques include sharp dissection, monopolar electrocautery, bipolar cautery and harmonic scalpel, while intracapsular techniques include microdebrider, bipolar radiofrequency ablation, or carbon dioxide laser.57

Mechanical stimulation of the endotracheal tube during extubation often results in laryngospasm which may induce type I-NPPE. Therefore, patient intubation should be carried out either under deep anesthesia or after muscle relaxant administration to lower the incidence of laryngospasms. Redundant controversies exist among anesthesiologists regarding which technique of tracheal extubation (awake versus anesthetized), is better at reducing the incidence of laryngospasms.58 With regards to this, Lee recommended extubation while the lungs are inflated by positive pressure.59 There is a close association between the type of surgery performed and the occurrence of laryngospasm. Tonsillectomy and adenoidectomy are associated with the highest incidence of laryngospasm (21-26%).58,60 However, induction of anesthesia in pediatrics either intravenously or by inhalation, using suitable anesthetic drugs is a controversial topic that should be addressed on a case by case basis. Intravenous anesthesia was found to be associated with a lower incidence of laryngospasm than inhalational anesthesia.61 Despite extensive debate regarding the avoidance of laryngospasm, several strategies have been suggested to lower the incidence of laryngospasm. Koc et al62 reported that topical lignocaine 2% at a dose of 4 mg/kg sprayed on the glottis, or 2% intravenous lignocaine given at 1 mg/kg five minutes before extubation are fairly effective in preventing laryngospasm following adenotonsillectomy. According to Gulhas et al,⁶³ 15 mg/kg magnesium sulfate in 30 ml 0.9% NaCl administered over 20 minutes after tracheal intubation has also been suggested to reduce the incidence of laryngospasm post-adenotonsillectomy. Intravenous administration of 0.25-0.5 mg/kg propofol prior to extubation has also been shown to be effective.58 Some anesthesiologists indicated that inhalation of 5% carbon dioxide could prevent laryngospasm, since respiratory drive to exhale the carbon dioxide overrides the laryngospasm reflex.

In conclusion, pATPE is a rare but life-threatening complication that requires immediate intervention. Young age (<3 years) or the presence of some comorbid conditions such as asthma, history of prematurity, craniofacial malformations, Down syndrome and severe OSA may increase the risk of developing PE in children undergoing adenotonsillectomy. With prompt diagnosis and intervention, most pATPE patients can be treated without incident. Since PE develops in a matter of minutes after adenotonsillectomy, simply monitoring oxygen saturation using pulse oximetry in the operating room and in the post-anesthesia care unit usually identifies patients at risk and in need of in-patient hospitalization rather than ICU admission.