Management of acute inflammatory childhood stridor

Abstract

BACKGROUND AND OBJECTIVE: Acute epiglottitis, bacterial tracheitis, and severe viral laryngotracheobronchitis continue to cause life-threatening airway obstruction in children. The aim of this retrospective study was to evaluate deficiencies in treatment, to clarify the role of airway endoscopy, and to identify current reasons for fatalities in these diseases.

MATERIALS AND PATIENTS: During the observation period from 1980 to 1996, we identified 61 patients treated in the pediatric intensive care unit of the University Hospital of Cologne for severe inflammatory airway obstruction.

RESULTS: Critical points in the appropriate airway management were (1) the confirmation of the admission diagnosis, (2) whether an artificial airway or rigid tracheobronchoscopy was required; and (3) appropriate timing of extubation. Since 1989 airway evaluation with flexible or small, rigid endoscopes was used to confirm the diagnosis and to determine the appropriate management. Endoscopic findings were extremely helpful to decide not to intubate and to monitor in the pediatric intensive care unit first. Three patients with acute epiglottitis died after out-of-hospital cardiorespiratory arrest.

CONCLUSION: Endoscopy should be a part of every established treatment protocol of childhood stridor. The most decisive factor to decrease mortality seems to be timely presentation to a referral center.

In the pediatric population, acute epiglottitis (AE), bacterial tracheitis (BT), and viral laryngotracheobronchitis (LTB) are frequent causes of central airway obstruction. The medical management of acute respiratory distress is a challenge, particularly in children with rapidly deteriorating symptoms and signs.

The onset of stridor in a young child usually represents LTB. In the management of this common disease, treatment with cool mist, nebulized racemic epinephrine, and/or a brief course of glucocorticosteroids is appropriate in most cases. 1 If standard treatment fails and further atypical deterioration occurs in LTB, an artificial airway must be considered, 2 and the initial diagnosis should be reviewed. AE, foreign-body aspiration, retropharyngeal abscess, laryngeal diphtheria, and particularly BT should be excluded by endoscopic or radiologic examination. 3 , 4

Bacterial croup was commonly discussed in pediatric textbooks until the 1950s. 5 It seemed to disappear as a clinical entity with the onset of antibiotic therapy and was “rediscovered” in 1979 by Han et al 6 and Jones et al, 7 the latter terming it bacterial tracheitis. The clinical distinction from other inflammatory airway is difficult. 8 , 9 A major advance in the management of BT was the introduction of rigid endoscopy for the removal of mucopurulent tracheal secretions. Along with nasotracheal intubation, it reduced the need for tracheostomies. 10

The treatment and outcome of patients with AE has changed considerably in the years since Sinclair 11 described the first case in the early 1940s. The recent literature reports a significant decrease in incidence after the establishment of the Haemophilus influenzae type B vaccine, 12 an increasing number of amoxicillin-resistant H influenzae infections, 13 an age shift toward younger children, 14 and a decreased number of complications in artificial airway management, favoring nasotracheal intubation over tracheostomy. 15

Nevertheless, AE, BT, and severe LTB continue to be potentially life-threatening conditions. 14 – 19 Management protocols and diagnostic procedures are topics of ongoing discussion, and in particular the indications for endoscopy are still unclear. 13 – 15 It was these reports that prompted this study at our institutions with the aim to evaluate deficiencies in diagnosis and management, to clarify the role of airway endoscopy, and to establish the causes of death.

SUBJECTS AND METHODS

The subjects of our retrospective study were all children treated in the pediatric intensive care unit (PICU) of the Children's Hospital (University of Cologne) between January 1, 1980, and December 31, 1996, who had LTB, BT, and AE. In total, 1509 children were hospitalized for acute obstructive inflammatory diseases of the central airway during this period. Most (96%, n = 1448) of these patients were admitted with the diagnosis of LTB and treated on the general ward. During this period 19 children were treated in the PICU for LTB, 16 for BT, and 26 for AE. The records of these 61 patients were reviewed.

Fig 1.

Classification of respiratory distress and initial therapeutic procedures in the emergency department. +Airway endoscopy has been performed additionally since 1989.

RESULTS

All children with respiratory distress were admitted to the emergency department of the Children's Hospital. The severity of the respiratory distress was staged according to the classification summarized in Fig 1. If AE or BT was suspected, or if the diagnosis remained unclear, an interdisciplinary consultation with members of the pediatric, otorhinolaryngologic, and anesthetic teams was organized.

Clinical Features on Admission

Most patients with AE had a typical combination of agitation, stridor, dysphagia, drooling, and occasionally cough. Clinical symptoms in most patients with BT and LTB included stridor and cough as typical signs of sub-glottic inflammation. Difficulties in differential diagnosis were particularly observed in BT, with 11 of 16 diagnoses being incorrect until an endoscopic examination was performed. Initial clinical findings are summarized in Table 1.

Airway Endoscopy and Management

Further diagnostic and therapeutic procedures depended on respiratory conditions at admission and are condensed in Fig 1. In patients with stage III or IV respiratory distress, an artificial airway was established immediately in the emergency department. Since 1989 diagnostic fiberoptic nasolaryngoscopy or indirect laryngoscopy with a small, rigid 70-degree endoscope was performed in cooperative, awake children (respiratory distress stage I-II). In uncooperative children, patients were transported to the operating theater for laryngotracheobronchoscopy under general anesthesia. The stabilized patient then was transferred to the PICU for further treatment.

Viral LTB. Intubation was immediately performed in 2 patients admitted in stage IV respiratory distress. A deterioration in the airway requiring nasotracheal intubation became necessary in 6 children with severe LTB 30 minutes to 5 hours after admission. Rigid tracheobronchoscopy was performed under general anesthesia in 3 of these patients to rule out BT, aspirated foreign bodies, and malformations of the tracheobronchial tree. The endoscopic findings were consistent with the typical LTB picture of an inflamed and edematous tracheal wall and absence of supraglottic inflammation and mucopurulent tracheal secretions.

Seven patients had fiberoptic nasolaryngoscopy and/or 70-degree laryngoscopy, permitting the exclusion of AE. The view through the vocal cords presented an inflamed subglottic region without thick tracheal secretions. Six of these patients were treated in the PICU without artificial airways. Improvement with standard therapy for LTB confirmed the final diagnosis in the remaining patients.

BT. Five patients were immediately intubated after admission because of impending total airway obstruction (respiratory distress stage III/IV). In the remaining patients, foreign-body aspiration was suspected in 3 children; standard treatment of LTB was first administered but failed to be effective in 7 children. These patients were transferred to the ENT operation theater for rigid tracheobronchoscopy. Endoscopic evaluation of the larynx and trachea revealed the typical features of BT: membranous inflammation or copious mucopurulent secretions in the trachea with notable subglottic edema and normal supraglottic structures. Subglottic membranous secretions were removed endoscopically, and patients were transported to the PICU after surgery.

In an 11-year-old girl fiberoptic nasolaryngoscopy was performed in the awake child, and the transglottic view revealed purulent tracheal secretions and absence of supraglottic inflammation. No artificial airway was required because airway obstruction was mild.

AE. Three children arrived in the emergency department in a very poor posthypoxic condition, and although cardiac arrest was reversible, brain death had already taken place. Intubation was performed in 16 patients admitted with AE, tracheostomy was required only in 1 girl after failure of intubation. One child had been intubated by the ambulance service before arriving at the hospital. Endotracheal tubes were smaller than usually estimated for the child's age. Direct laryngoscopy during intubation revealed various degrees of inflammation of the supraglottic structures involving the spaces between the arytenoids and the epiglottis, confirming the diagnosis in these children.

In 7 children, conservative management in the PICU was successful without the establishment of an artificial airway. Direct pharyngoscopy with the aid of a tongue depressor was performed in patients with moderate respiratory distress (stage I-II) without complications. Since 1989 the decision against intubation was based on clinical appearance and the visualization of the supra-glottitis. Fiberoptic nasolaryngoscopy and/or indirect laryngoscopy with a rigid 70-degree endoscope was performed in 6 cooperative older children with AE (mean age 6 years).

Therapy for nonintubated patients with LTB (respiratory conditions stage I-II) included humidified air/oxygen and nebulized racemic epinephrine inhalations and oral or intravenous glucocorticosteroids. Additionally, ampicillin (150 mg/kg) was administered when the body temperature increased during hospitalization to prevent a possible progression toward bacterial superinfection in nonintubated children or in all patients with artificial airways.

Table 1.

Initial clinical findings

	Diagnosis
Finding	LTB (n = 19)	BT (n = 16)	AE (n = 26)
Age (y)
Mean	4.3	5.2	3.9
Range	0.75–9.5	0.75–11.5	2–10
Sex (M/F)	15/4	12/4	17/9
Prodromal period (range)	0–14 d	0–48 h	0–48 h
Up to 6 h (n)	7	0	5
Up to 12 h (n)	5	8	13
Up to 2 d (n)	7	9	8
Symptoms on admission (%)
Cough	83	67	12
Dysphagia	28	36	62
Hoarseness	24	43	21
Drooling	11	8	58
Inflammatory parameters
Mean body temperature (°C)	38.3	38.5	39.0
Body temperature > 38.5 (n)	5	6	16
Mean WBC (cells/nL)	10.4	9.9	16.9
Admitting diagnoses (n)
Epiglottitis	2	2	21
BT	1	5	0
Viral LTB	12	7	3
Aspiration	3	2	1
Pneumonia	1	0	1
Chest x-ray findings (n)	9	14	16
Bronchopneumonia	2	3	9
Bronchitis	5	5	0
Lobar atelectasis	0	2	2
No pneumonia	2	4	6

WBC, White blood cell count.

With AE, intravenous antimicrobial therapy was a combination of ampicillin (150 mg/kg) and chloramphenicol (50 mg/kg) in the early 1980s, and later cefotaxime (50 mg/kg). All children with BT received parenteral antibiotic treatment with either ampicillin (150 mg/kg) and dicloxacillin (100 mg/kg) or piperacillin (100 mg/kg) and tobramycin (5 mg/kg).

Microbiologic Findings

In AE, blood cultures were positive in 62% of examinations and always demonstrated ampicillin-sensitive H influenzae type b. The cultures of the purulent tracheal secretions of BT showed Staphylococcus aureus in combination with various other pathogens in 6 children; the involvement of Pseudomonas aeruginosa was noted in 3 cases.

Table 2.

Airway management: Endoscopic examinations and artificial airways

	1980–1988			1989–1996
	LTB (n = 10)	BT (n = 9)	AE (n = 13)	LTB (n = 9)	BT (n = 7)	AE (n = 13)
Respiratory distress on admission
Stage I	1	0	2	1	1	2
Stage II	5	6	6	5	3	8
Stage III	3	2	2	2	2	2
Stage IV	1	1	3	1	1	1
Airway examination
Direct pharyngoscopy	6	6	8	6	5	10
Fiberoptical nasolaryngoscopy	0	0	0	6	1	5
Indirect laryngoscopy	0	0	0	4	0	4
Direct laryngoscopy (during intubation)	5	9	12	2	6	6
Rigid tracheobronchoscopy	2	9	0	1	6	0
Laryngo-/tracheoscopy (before extubation)	0	1	10	1	5	6
Airway management
No. monitored without artificial airway (%)	5 (50)	0	1 (8)	6 (67)	1 (14)	6 (46)
No. of intubations (%)	5 (50)	9 (100)	11 (84)	3 (33)	6 (86)	7 (54)
Mean duration of intubation (range)	48 h	5 d	62 h	48 h	4d	54 h
	(1–4 d)	(2–16 d)	(12 h-6 d)	(1–4 d)	(1 h-6 d)	(1 h-5 d)
No. of reintubations (%)	0	3 (33)	0	0	1 (14)	0
No. of tracheostomies (%)	0	0	1 (8)	0	0	0
No. dead of disease	0	0	2	0	0	1

Because of multiple airway examinations performed in some patients, the total number may be higher than the number of patients.

Outcome, Complications, and Mortality

The usual course was slow resolution of the inflammatory process, and the period of hospitalization varied between 1 and 3 weeks. The timing of elective extubation was based on the absence of fever (body temperature < 38°C) for more than 12 hours, increasing leakage around the endotracheal tube indicating reduced inflammation, diminishing quantity of secretions suctioned from the endotracheal tube, and later on, endoscopic findings. Patients with BT required the longest duration of endotracheal intubation. Severe respiratory failure resembling acute respiratory distress syndrome developed in 1 child with BT, and the patient required mechanical ventilation for 16 days. Recurrence of severe upper airway obstruction necessitating reintubation occurred in 4 children with BT. Flexible tracheoscopy through the endotracheal tube allowed determination of the exact timing of extubation in 6 children with BT.

No failure of elective extubation was noted in the patients with AE and LTB. The decision for extubation was based on clinical recovery, and the supraglottis was reexamined with a MacIntosch laryngoscope if leakage around the artificial airway occurred in AE. Mortality occurred only from AE (11.5%). In the remaining patients, no major complications were recognized. Data of airway management are condensed in Table 2.

DISCUSSION

Diagnosis and Airway Management

Modern treatment protocols have resulted in a marked decrease in long-term morbidity and mortality in severe upper airway inflammation. 10 , 13 – 18 The material of this study identifies 3 critical points in the management endangering patient outcome.

The first point was confirmation of the diagnosis on admission. Diagnostic difficulties were noted in 3 children with AE who had croupy coughs at presentation, in 2 severe cases of LTB admitted with dysphagia and drooling, and in most patients with BT demonstrating clinical signs similar to those of LTB. Therefore it was not unexpected that the initial diagnosis in children with BT was frequently found to be LTB, foreign-body aspiration, or AE in this series and in others. 8 , 9 , 17

Mauro et al 20 were able to confirm all 3 cases of AE in 155 children admitted with acute stridor; the diagnosis was based on symptoms and direct pharyngoscopy. In contrast to our investigation, Mauro et al did not recognize cough as an atypical sign in these 3 patients. In our opinion direct pharyngoscopy is not sufficient because it is not always possible in awake children. Moreover, this method is not capable of detecting sub-glottic lesions, such as BT, which was not seen at all in the study of Mauro et al.

Some authors recommend plain lateral neck radiographs to confirm a diagnosis of AE 13 , 14 , 21 or BT. 6 , 8 , 16 Typical neck radiographs of patients with AE reveal inflammatory thickening of supraglottic structures. However, the absence of these findings is not sufficient to exclude AE, particularly in the early stages of the disease.

Like others, 1 we think that direct or endoscopic inspection of the epiglottis is indispensable. In BT, most radiologic findings were characterized as subglottic narrowing, a feature known to also appear in LTB. 9 If an airway emergency is present, we agree with Dudin et al 22 that lateral neck radiographs “… are of pure academic interest.” Therefore we usually refrained from obtaining lateral neck radiographs, even in stable patients. In contrast, additional information was gained from chest x-ray films, and high rates of bronchopulmonary involvement were noted in BT (71%), AE (52%), and LTB (78%) in our study (Table 1), as in most other studies. 6 , 7 , 9

As soon as the initial diagnosis is confirmed, an artificial airway must be considered. The current literature recommends intubation or tracheostomy in every patient with AE or BT, 1 , 9 although several studies refer to successful management without artificial air-ways. 6 – 8 , 13 , 15 – 19 , 23 – 25 Besides the clinical picture, airway endoscopy was the key to deciding whether an artificial airway was needed in patients with LTB with moderate respiratory distress (stage I-II).

Endoscopic airway evaluation was indispensable in our patients, as in the vast majority of children with BT reported in recent studies. 10 , 16 , 18 , 26 If BT is suspected, fiberoptic nasolaryngoscopy can be performed in awake children with mild airway obstruction to rule out epiglottitis, and occasionally this may be the only endoscopic management necessary. In clinically severe airway obstruction (respiratory distress stage III-IV), it should be always performed with the patient under general anesthesia, and rigid endoscopes should be used before intubation for cautious suctioning of secretions from the trachea. This also permits the exclusion of other causes, such as foreign-body aspiration or severe LTB. 10

Indirect laryngoscopy is an established diagnostic tool in adults with AE but not in children, probably because of the fear of total airway obstruction after physical examination. 20 , 27 , 28 In this series fiberoptic nasolaryngoscopy or endoscopy with small, rigid laryngoscopes depicted detailed information about the airway conditions without complications. The endoscopic findings were extremely helpful in deciding not to intubate and to monitor in the PICU first. The number of patients treated with artificial airways has decreased from 92% to 54% since endoscopy became available in 1989. Therefore we agree with Andreassen et al 15 that endoscopy should be performed in awake children with AE, particularly considering the up to 30% airway complication rate in artificial ventilation. 16 , 23 In our series the target group for fiberoptic examination was cooperative older children (mean age 6 years), with moderate respiratory distress (stage I-II) at presentation.

Nevertheless, if total airway obstruction is a threat, nasotracheal intubation should be performed immediately. In this series, failure of intubation occurred in only 1 child with AE, who underwent emergency tracheostomy immediately after admission (4%). Although the literature favors nasotracheal intubation as the method of choice for airway maintenance, tracheostomy is still in use in up to 14% of patients with AE. 14

The third critical point in the treatment of our patients was appropriate extubation based on clinical improvement. Recurrence of respiratory distress required reintubation and further rigid endoscopy in 25% of patients with BT. No failure of extubation occurred in patients with supraglottic inflammation.

Microbiology

In BT, S aureus was the most frequently recognized microorganism observed in monocultures in most previous series. 7 – 9 , 16 Besides S aureus, we recognized the involvement of P aeruginosa in 2 patients, and this should be considered when choosing the initial antibiotic therapy.

In contrast to the findings of current North American investigations reporting increasing numbers of ampicillin-resistant H influenzae organisms in AE (Kessler et al 13 and Emmerson et al 14 ), we did not note any evidence of β-lactamase–producing strains.

Outcome and Mortality

In a review of the literature, mortality from BT occurred as a result of out-of-hospital cardiopulmonary arrest with irreversible neurologic deficits 9 , 16 , 18 and as a result of ineffective airway clearance during initial intubation. 26 No mortality or significant long-term morbidity was recognized in our patients and, in particular, no tracheal stenosis. This again emphasizes the importance of rigid tracheoscopy and suctioning before intubation.

One remarkable finding of our study was the high mortality rate (11.5%) in AE. Life-threatening states originated from out-of-hospital cardiorespiratory arrest, resulting in irreversible posthypoxia neurologic deficits. Similar reasons were cited in the reviewed studies, with the reported mortality varying between 0% and 2%. 13 – 15 , 19 , 20 , 23 Baker and Ludwig 29 in 1991 reviewed the methods of transport of seriously ill children from the physician's office to a referral center in Philadelphia. Even if most physicians had professional services available, more than half of them indicated that the family automobile was the most useful means of transporting ill children to tertiary care centers, including patients with suspected epiglottitis. We can conclude from our data that adequate community management of childhood stridor is not practiced and that the most decisive factor to decrease mortality in AE is the establishment of an appropriate out-of-hospital management protocol. If AE or BT is suspected, if LTB appears severe, or if the diagnosis of acute stridor remains unclear, the child should be immediately transported to an appropriate center by use of professional ambulance services.

CONCLUSION

The data presented in this article reflect the current clinical practice at our institution and represent our current philosophy concerning the role of airway endoscopy in acute childhood stridor. We believe that fiberoptic airway endoscopy represents a major step forward in the management of acute obstructive airway disease and that this requires close interdisciplinary collaboration between pediatricians, anesthetists, and otorhinolaryngologists. Airway endoscopy should be integrated in current management protocols of childhood stridor. However, this clinical experience may not reflect the mainstream of established management protocols but may be of special relevance for those centers in which no formal pediatric otolaryngology team is involved with the primary care of the child.

References

Cressman

Myer

CM.

Diagnosis and management of croup and epiglottitis. Pediatr Clin North Am 1994; 41:265–76.

Wilson

DF.

Inflammatory disease of the airway. In: Meyer

III Cotton

Shott

editors. The pediatric airway. Philadelphia: WB Saunders; 1995. p. 67–99.

Manning

Ridenour

Brown

Measles: an epidemic of upper airway obstruction. Otolaryngol Head Neck Surg 1991; 105:415–7.

Stern

Boat

TF.

Lower respiratory tract. In: Behrman

Kliegman

Vaughan

editors. Nelson textbook of pediatrics. 14th ed. Philadelphia: WB Saunders; 1992. p. 1064–8.

Nelson

WE.

Bacterial tracheitis: a historical perspective. J Pediatr 1984; 105:52–5.

Han

Dunbar

Striker

TW.

Membranous laryngotracheobronchitis (membranous croup). AJR Am J Roentgenol 1979; 133:53–8.

Jones

Santos

Overall

JC.

Bacterial tracheitis. JAMA 1979; 242:721–6.

Friedman

Jorgensen

Healy

Bacterial tracheitis—two year experience. Laryngoscope 1985; 95:9–11.

Sofer

Duncan

Chernick

Bacterial tracheitis, an old disease rediscovered. Clin Pediatr (Phila) 1983; 22:407–11.

10.

Eckel

Widemann

Damm

Airway endoscopy in the diagnosis and treatment of bacterial tracheitis in children. Int J Pediatr Otorhinolaryngol 1993; 27:147–57.

11.

Sinclair

SE.

Haemophilus influenzae type b in acute laryngitis with bacteriemia. JAMA 1941; 117:170–3.

12.

Committee on Infectious Diseases. American Academy of Pediatrics. Haemophilus influenzae type b polysaccharide vaccine. Pediatrics 1985; 76:322–4.

13.

Kessler

Wetmore

Marsh

RR.

Childhood epiglottitis in recent years. Int J Pediatr Otorhinolaryngol 1993; 25:155–62.

14.

Emmerson

SGP

Richman

Spahn

Changing pattern of epiglottitis in children. Otolaryngol Head Neck Surg 1991; 104:287–92.

15.

Andreassen

Baer

Nielsen

Acute epiglottitis—25 years experience with nasotracheal intubation, current management policy and future trends. J Laryngol Otol 1992; 106:1072–5.

16.

Liston

Gehrz

Siegel

Bacterial tracheitis. Am J Dis Child 1983; 137:764–7.

17.

Tan

AKW

Manoukian

JJ.

Hospitalized croup (bacterial and viral): the role of rigid endoscopy. J Otolaryngol 1992; 21:48–53.

18.

Kasian

Bingham

Steinberg

Bacterial tracheitis in children. Can Med Assoc J 1989; 140:46–50.

19.

Trollfors

Nylen

Strangert

Acute epiglottitis in children and adults in Sweden 1981–3. Arch Dis Child 1990; 65:491–4.

20.

Mauro

Poole

Lockhard

CH.

Differentiation of epiglottitis from laryngotracheitis in the child with stridor. Am J Dis Child 1988; 142:679–82.

21.

Vetto

RR.

Epiglottitis: report of 37 cases. JAMA 1960; 17:990–4.

22.

Dudin

Thalji

Rambaud-Cousson

Bacterial tracheitis among children hospitalized for severe obstructive dyspnea. Pediatr Infect Dis J 1990; 9:293–5.

23.

Damm

Eckel

Jungehülsing

Airway endoscopy in the interdisciplinary management of acute epiglottitis. Int J Pediatr Otorhinolaryngol 1996; 38:41–51.

24.

Wetmore

Handler

SD.

Epiglottitis: evolution in management during the last decade. Ann Otol Rhinol Laryngol 1979; 88:822–6.

25.

Damm

Eckel

Roth

Interdisciplinary therapeutic concepts in severe bacterial infections of the central respiratory tract in childhood. Laryngorhinootologie 1996; 75:293–300.

26.

Donnelly

McMillan

Weiner

LB.

Bacterial tracheitis, report of eight new cases and review. Rev Infect Dis 1990; 12:729–35.

27.

Cox

Bates

Drake-Lee

The use of flexible nasolaryngoscopy in adults with acute epiglottitis. Ann R Coll Surg Engl 1988; 70:361–2.

28.

Cherry

JD.

Invasive Haemophilus infections: epiglottitis and cellulitis. Ann Otol Rhinol Laryngol 1981; 90:19–22.

29.

Baker

Ludwig

Pediatric emergency transport and the private practitioner. Pediatrics 1991; 88:691–5.