Therapeutic hypothermia is an established standard of care in the treatment of hypoxic-ischemic encephalopathy. Application of therapeutic hypothermia in the clinical setting may reveal a wider spectrum of adverse events than previously reported. We report 5 cases of transient respiratory stridor in 51 infants, occurring at different time points in the cooling process, which appeared to be unrelated to the intubation procedure. Therapeutic hypothermia was associated with transient stridor in this case series. Formal laryngoscopy is required to determine the underlying pathologic etiology.
- ETT —
- endotracheal tube
- HIE —
- hypoxic-ischaemic encephalopathy
In resource-rich countries, therapeutic hypothermia is a standard of care in the treatment of moderate and severe neonatal hypoxic-ischemic encephalopathy (HIE).1 Several meta-analyses, involving >1500 infants, have revealed no serious adverse effects of therapeutic hypothermia.2 Nevertheless, professional recommendations are that treatment should occur in ICUs with sufficient expertise to manage such neurocritical infants.3,4 Cooling registries of prospectively collected patient information are maintained to monitor clinical practice and to report on rare side effects that may not have been apparent in the context of clinical trials.5,6
Intubation and ventilation are commonly undertaken interventions in asphyxiated infants, but the duration of support is usually short and limited to the period of acute encephalopathy.5 Cooling may increase oxygen requirements acutely but has not been otherwise reported to alter the respiratory course of asphyxiated infants.7 We present a series of 5 asphyxiated infants who underwent therapeutic hypothermia, and who developed stridor during their clinical course.
Cases of stridor were identified from all term (>36 weeks’ gestation) admissions in a 4-year period from January 1, 2009, until December 31, 2012. Electronic records of infants with a diagnosis of stridor at admission, at discharge, or during hospital stay were reviewed.
There were 26 413 term infants born in the study period, with an incidence of term HIE of 1.7 per 1000 live births. Fifty-one asphyxiated inborn and outborn infants underwent therapeutic hypothermia over the study period.
Incidence of Stridor in Term Infants
Twenty-five term infants were identified with stridor either before (n = 19) or after (n = 6) admission to the neonatal unit, an incidence of 0.9 per 1000 live births. Of those with stridor on admission, the diagnosis was a congenital abnormality (n = 11), vocal cord palsy (n = 4), and a neurologic disorder (n = 1); and in 3 infants the cause was unidentified but symptoms were transient and mild. Five cases occurred after admission in infants undergoing therapeutic hypothermia, 9.6% of all cooled infants; each is described below (Table 1). Only 1 nonasphyxiated term infant had transient stridor reported after an intubation period of 20 minutes.
Cooling and Intubation Protocols
The local cooling protocol is based on standard recommendations for the treatment of asphyxiated newborns.8,9 Whole-body hypothermia is induced and maintained by using the CritiCool Cooling System (Charter Kontron, Milton Keynes, United Kingdom). Sedation of infants with morphine is standard practice. Endotracheal intubation is performed when clinically indicated, and infants may be extubated during therapeutic hypothermia if status allows. Junior staff are encouraged to attempt 2 intubations under senior supervision. The use of a stylet introducer is not routine and is discouraged. An uncuffed, nonshouldered endotracheal tube (ETT) is used, and placement is confirmed with a carbon dioxide colorimeter, auscultation, and chest radiograph. Endotracheal suction is performed when clinically indicated, although not routinely. Inspired ventilator gases are humidified and heated to 37°C.
Cases of Stridor in Infants Undergoing Therapeutic Hypothermia
Five infants (1 boy, 4 girls) were born between 39 and 41 weeks’ gestation, with birth weights from 2770 to 4240 g. All infants suffered intrapartum asphyxia, and arterial cord pH values ranged from 6.86 to 6.89. No infant experienced dystocia or showed signs of peripheral nerve paralysis during admission. All infants developed moderate/severe HIE and underwent therapeutic hypothermia. There was evidence of mild renal and hepatic involvement in 3 infants (Table 1).
At the time of stridor, serum calcium and magnesium levels were normal in 3 infants, 1 infant had an ionized calcium level of 0.88 mmol/L, and another had a magnesium level of 0.69 mmol/L. Serum sodium ranged from 125 to 134mmol/L, and the urine output range was 1.1 to 5.1 mL/kg per hour at the time of stridor, reflecting the variation in output seen in acute renal injury and recovery. No infant showed interstitial edema other than that occurring normally during the cooling process; there was also no evidence of significant fluid overload. In only 1 infant (case 1), the larynx was inspected at reintubation when there was no report of obvious airway abnormality.
At birth, meconium was suctioned from this infant’s trachea. Intubation was successful on the third attempt with an introducer and a size 3.5 ETT at 4 minutes. The infant was extubated at 10 hours into air with good respiratory drive.
Twelve hours into rewarming, with a core temperature of 36.5°C, marked respiratory distress was observed with an obstructive breathing pattern and respiratory acidosis (Pco2 of 10.3 kPa). Uneventful reintubation under direct laryngoscopy relieved the work of breathing, and despite minimal oxygen requirement, surfactant was administered because of suspected hypothermia-induced surfactant deficiency. The infant was not receiving morphine or anticonvulsants and was extubated to air 11 hours later with no further respiratory difficulty.
At birth, the infant received mask ventilation and cardiac compressions. Regular respiratory effort was established at 12 minutes. From 1 hour of age the infant was nursed in air and was never ventilated, nor was intubation or endotracheal suction ever attempted.
Four hours into rewarming, with a core temperature of 35.1°C, acute stridor was noted with tracheal tug and desaturation (Pco2 of 8.5 kPa). Airway obstruction was ameliorated with a nasopharyngeal airway. A morphine infusion was reduced and rewarming slowed to 0.2°C per hour. Antibiotics were commenced for possible bacterial tracheitis, but infection markers and cultures proved negative. The stridor resolved within 48 hours.
At birth, the infant received mask ventilation and chest compressions. Intubation occurred at 7 minutes with a size 3.5 ETT by the senior doctor after 1 attempt by the junior doctor. One dose of phenobarbitone was administered for seizures at 13 hours.
At 55 hours of age, with a core temperature of 33.5°C, the infant was extubated to air while receiving morphine. Mild biphasic stridor was noted 4 hours later but was not associated with increased work of breathing nor did it require any intervention (Pco2 of 5.9 kPa). Stridor was evident for a further 24 hours, and the subsequent rewarming period was uncomplicated.
At birth, the infant received mask ventilation and cardiac compressions. Intubation occurred on the first attempt at 7 minutes with a size 3.5 ETT. On day 1, there was an accidental extubation and reintubation occurred on the second attempt with a size 4 ETT. Two doses of phenobarbitone were administered for seizures.
On day 6 of life, 36 hours after reaching normothermia, extubation occurred. There was immediate stridor, increased work of breathing, desaturation, and the Pco2 increased to 7.1 kPa. Symptoms were relieved with nebulized adrenaline, intravenous dexamethasone, and continuous positive airway pressure; and intermittent mild stridor was audible for an additional 30 hours.
At birth, suction was performed under direct vision. Intubation occurred on the second attempt at 7 minutes with a size 3.5 ETT. The infant was extubated at 6 hours of age and remained breathing in air through treatment with phenobarbitone and phenytoin for seizures at 14 hours of age. Rewarming followed uneventfully at 72 hours of age.
Seven hours after reaching normothermia, there was desaturation and increased work of breathing with biphasic stridor (Pco2 of 8.3 kPa). The infant was nursed prone and placed on continuous positive airway pressure for 24 hours until stridor had resolved.
As widespread implementation of therapeutic hypothermia occurs, regional guidelines and pathways have been increasingly established and early concerns about the safety of therapeutic hypothermia are subsiding as a body of experience accrues within the neonatal community.
However, although randomized trials and meta-analyses reassure that significant side effects of cooling are few and mild, they are not powered to detect rare events that may become apparent when therapeutic hypothermia is implemented on a wider scale and within routine practice. In addition, rarely is the scope of randomized trials such that all possible adverse events are sought or evaluated.
National registers encourage reporting of individual cases to audit practice and outcomes but also to inform on the rare side effects of cooling.5,6 Of relevance is the evidence that subcutaneous fat necrosis and lidocaine-induced arrhythmias may be more common in infants who undergo therapeutic hypothermia.5,10
We report on 5 infants who underwent therapeutic hypothermia for intrapartum asphyxia and who developed stridor during admission. Stridor was not necessarily related to tracheal intubation, difficulty in intubation, or the duration of ventilation. Intubation practice did not change during the study period nor was stridor observed in noncooled, ventilated infants, suggesting that this phenomenon was not related to the procedure of intubation alone. Moreover, stridor was not temporally related to extubation and occurred as remotely as 71 hours after ETT removal. A limitation of this series is that formal laryngeal examination was undertaken in only 1 case, but the temporary nature of the obstruction implies a functional rather than a structural etiology. Nevertheless, direct laryngoscopy is essential in determining pathologic etiology and is planned in future cases.
Stridor occurred during all phases of hypothermia treatment, including during active cooling, rewarming, and normothermia. In all cases, the stridor was transient and resolved within 48 hours. All infants demonstrated good respiratory drive, and the absence of encephalopathy or sedation at the time of stridor in some cases implies that neurologic depression or injury is unlikely to have contributed to the acute airway obstruction observed.
It is well recognized that glottal mechanics are sensitive to changes in core temperature, such that increases in temperature result in an increase in the force of the glottal closing reflex and vice versa.11 In animals, the glottal closing force increases 300% to 400% over a similar temperature range as occurs during rewarming,12 and it may follow that temperature fluctuations experienced during servo-controlled cooling result in alteration of glottal mechanics. Alternatively, the increase in the extracellular fluid compartment secondary to capillary leak that occurs during cooling13 has the potential to result in edema of the upper airway and an increase in resistance, which resolves as this compartment contracts.
No cases of respiratory stridor have been reported to the UK TOBY Cooling Register or the Vermont Oxford Network (VON) Neonatal Encephalopathy Registry (D. Azzopardi, FRCHPCH and B. Strohm, RN on behalf of the UK TOBY Cooling Register, personal communication, 2013; R. Pfister, MD on behalf of the VON Neonatal Encephalopathy Registry, personal communication, 2013), although the authors are aware of 2 infants in 2 other UK institutions who developed stridor during therapeutic hypothermia. We assume that occasional instances may not be regarded as an association of therapeutic hypothermia per se but as a complication of intubation, and therefore the problem may be underreported. In addition, almost half of all centers providing data to the TOBY register reported that they treat ≤6 infants per year, and as such, infrequent adverse effects may not have been experienced (D. Azzopardi and B. Strohm on behalf of the UK TOBY Cooling Register, personal communication, 2013).
Although not described as a significant adverse effect of cooling in meta-analyses, a multicenter safety pilot trial reported the occurrence of stridor in 9 of 31 cooled infants compared with 1 of 33 control infants (P < .01), an incidence of 29%.14 These infants responded to nebulized adrenaline, and the authors hypothesized that the stridor was related to transient tracheal swelling after ventilation with humidified air of 34°C. In our center, inspired gases are warmed and humidified to 37°C.
Therapeutic hypothermia has been shown to improve outcome in adults after out-of-hospital cardiac arrest; stridor or airway obstruction is not reported as an adverse effect.15,16 Published cohort studies in pediatric patients undergoing hypothermia after cardiac arrest do not list stridor as an adverse effect,17,18 but the outcomes of a large randomized controlled trial in this population have yet to be published (F. Moler, MD, MS, FCCM on behalf of the Therapeutic Hypothermia after Pediatric Cardiac Arrest- Out of Hospital (THAPCA-OH) trial, personal communication, 2013).
In our case series, stridor was associated with hypothermia treatment in hypoxic-ischemic encephalopathy, but not with severe compromise. In all infants, symptoms resolved within 48 hours. The neuroprotective benefit of therapeutic hypothermia outweighs this small risk of transient stridor.
- Accepted December 26, 2013.
- Address correspondence to Julie-Clare Becher, FRCPCH, MD, FRCPE, MB, ChB, Department of Neonatology, Simpson Centre for Reproductive Health, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom EH16 4SA. E-mail:
Dr Orme participated in the concept and design, collected and analyzed data, and drafted the initial manuscript; Dr Kissack participated in the concept and design, designed the data collection instrument, coordinated data collection and analysis, and drafted the initial manuscript; Dr Becher conceptualized and designed the study, collected and analyzed data, and reviewed and revised the initial manuscript; and all authors approved the final manuscript as submitted.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- ↵Jacobs SE, Berg M, Hunt R, Tarnow-Mordi WO, Inder TE, Davis PG. Cooling for newborns with hypoxic ischaemic encephalopathy. Cochrane Rev. 2013 (Jan 31)
- ↵National Institute for Health and Clinical Excellence. IPG347: therapeutic hypothermia with intracorporeal temperature monitoring for hypoxic perinatal brain injury. NICE Interventional Procedure Guidance, May 2010. Manchester, United Kingdom: National Institute for Health and Clinical Excellence; 2010
- ↵Adams E, Austin T, Becher JC, et al; British Association of Perinatal Medicine. Position statement on therapeutic cooling for neonatal encephalopathy. BAPM 2010
- Thoresen M,
- Whitelaw A
- ↵National Perinatal Epidemiology Unit. The UK TOBY Cooling Register. Available at: www.npeu.ox.ac.uk/tobyregister. Accessed November 1, 2013
- Strohm B,
- Hobson A,
- Brocklehurst P,
- Edwards AD,
- Azzopardi D,
- UK TOBY Cooling Register
- Hammersborg SM,
- Farstad M,
- Haugen O,
- Kvalheim V,
- Onarheim H,
- Husby P
- Nolan JP,
- Morley PT,
- Vanden Hoek TL,
- et al.,
- International Liaison Committee on Resuscitation
- Copyright © 2014 by the American Academy of Pediatrics