Abstract

Objective

The incidence of postoperative cough (primary outcome) and adverse respiratory events (secondary outcome) in children who received anesthesia by laryngeal mask airway (LMA) with children who received anesthesia by face mask (FM) was compared in a blind randomized trial with uncomplicated upper respiratory track infection (URI) undergoing general anesthesia. Previous studies of pediatric patients with URI receiving anesthesia by endotracheal tube have reported a greater number of anesthetic complications; however reports concerning adverse effects in pediatric patients with URI receiving anesthesia by LMA or FM are scanty.

Method

For the present trial, 150 children with uncomplicated URI and requiring general anesthesia for ophthalmic procedures were enrolled. Once the severity of preoperative URI symptoms was stratified, the children were randomized to receive general anesthesia by FM or LMA. Anesthesia was induced with sevoflurane and nitrous oxide in oxygen (N₂O in O₂). Respiratory adverse events were evaluated peri- and post-operatively.

Results

The two groups did not differ in age, weight, American Society of Anesthesiologists (ASA) physical status, sex, duration of surgery or severity of URI symptoms. The incidences of cough (19% in LMA vs. 42% in FM), vomiting (4% in LMA vs. 12% in FM) and intervention to maintain the patency of the airway were statistically higher in the FM group (p < 0.05). There were no differences between the two groups with respect to the incidences of apnea, laryngospasm, desaturation, bronchospasm, readmission and sore throat.

Conclusion

In children with uncomplicated URI, the administration of inhalation anesthetics in general anesthesia by LMA is likely to cause fewer adverse events than the use of FM.

Keywords

anesthesia; child; face masks; laryngeal masks; respiratory tract infections;

---

1. Introduction

General anesthesia (GA) in children with existing upper respiratory infection (URI) has been a subject of debate for many years.1, 2, 3, 4 There is a host of evidences to show that the hyperactivity of the airway following URI increases the risk of perioperative respiratory complications, including cough, laryngospasm, desaturation and bronchospasm.1, 2, 3, 4, 5, 6, 7 The laryngeal mask airway (LMA), which was commercially introduced as a supraglottic device in 1988, apparently induces less laryngeal stimulation.¹ It offers an attractive alternative for airway management in children with URI,1, 2, 3, 4 and the superiority of LMA over the use of an endotracheal tube (ETT) has been studied in children with URI.¹ Some reports assumed that the incidence of respiratory complications among children with existing URI following GA with LMA would be higher than that following FM;1, 2, 5 however, some studies showed that it made no difference.⁴ Therefore we designed a systemic randomized trial to compare the adverse events related to the use of LMA and FM for general anesthesia in paediatric patients with uncomplicated URI at the time of operation. We wished to clarify whether, for general anesthesia in children with existing uncomplicated URI, there were significant differences between LMA and FM with respect to primary (cough) and secondary (laryngospasm, bronchospasm and desaturation) outcomes.

2. Materials and methods

This study was approved by the Ethics Committee of Shahid Beheshti University of Medical Sciences. Informed consent was obtained from the parents or legal guardians of the children recruited to the study (in line with the British Medical Journal guidelines), who required elective ophthalmic examination under general anesthesia because of ophthalmic problems (e.g., glaucoma). The parents or legal guardians were free to refuse to consent to their child taking part in the study. Inclusion criteria included: age 1–12 years with existing uncomplicated URI, of American Society of Anesthesiologists (ASA) physical status class II. Children with intraocular infection or those with co-existing pulmonary disease, asthma, a cardiac problem or an abnormal airway, or undergoing procedures that needed muscle relaxant or intravenous induction, and those who required any additional medication, that is, an opioid or atropine, were excluded from the study. Also excluded were children with a full stomach, lower respiratory tract infection, oral temperature greater than 38°C, or who were unwell based on a parental statement or medical judgment (such as patients with purulent sputum).

One hundred and fifty children consecutively admitted during the period November 2009 to March 2010 to Labbafinejad Specialist Eye and Urology Hospital, Tehran, Iran, who met the acceptance criteria were enrolled in the study. They were stratified into two groups according to the severity of their URI. They were then randomized to receive general anesthesia by LMA (Well Lead Medical Co., Guangdong, China) insertion or by FM (Lifeline Systems Pvt. Ltd, New Delhi, India).

On the day of surgery, detailed information regarding the severity and onset of URI was recorded. A data sheet was given to each patient for preoperative information. The diagnosis of a URI was based on symptoms (cough, sneeze, runny nose, nasal congestion), and the child’s parent(s) or guardians(s) asked whether they thought the child had a cold that day.1, 2 URI symptoms, including, runny nose, cough, nasal congestion, sneezing, were scored from 1 to 4 according to their severity (Appendix 1).⁴ Parent(s) or guardians(s) were asked the onset time of the URI symptoms and whether the child was a passive smoker at home. Patients were stratified into two groups (mild and moderate) according to the severity of URI symptoms. Children who had at least two symptoms preoperatively and scored two or more were judged to have a moderate URI (Appendix 1).

Patients were then randomly allocated (using a table of random numbers) to receive general anesthesia conveyed by LMA or FM. The anesthesiologist performing the anesthesia was unaware of the severity group to which the patient belonged; the patient was blind to the way the anesthesia was given; and the postoperative symptom-assessors were also blind to the way the anesthesia was given to the patient. The anesthesiologist who examined the patient preoperatively also attended at the operating room while the on-duty anesthesiologist (semi-blind to patient’s severity of URI) was anesthetizing the patient. Both of them were present until the end of surgery and were available in the recovery room; another person recorded the condition of the patients in the recovery room. The three anesthesiologists who were entrusted with conducting the GA were briefed on the methodology and study protocol to increase the validity of the study results; they did not participate in the analysis of the results.

In both groups, anesthesia was induced with sevoflurane and nitrous oxide (N₂O) in 50% oxygen (O₂). The children were not given muscle relaxant. The only parenteral drug used after induction of anesthesia was intravenous lidocaine 1.5 mg/kg (preservative free lidocaine hydrochloride Lignodic 2%) (Caspian Tamin Co., Tehran, Iran). In the LMA group the device was inserted 3 min after injection of lidocaine, when the cerebral status index (CSI) reached between 40 and 60 (Cerebral State Monitor, Danmeter-Goalwick Holdings Ltd, Odense, Denmark) and the end expiratory sevoflurane concentration was at least 2.1 %. LMA size was selected according to the weight of the child, in line with the manufacturer’s instructions. Anesthesia was maintained with 2–3 % sevoflurane and N₂O in 50% O₂. All children breathed spontaneously without the aid of positive pressure ventilation. Monitoring parameters included pulse oximetry (SPO₂), electrocardiography, respiratory rate, end-tidal CO₂, manual blood pressure and CSI. Once the ophthalmic procedure was completed, N₂O was disconnected and the LMA (with inflated cuff) removed, with reduction of the sevoflurane dose. The opportune moment at which to remove the LMA was defined as: (1) breathing 100% oxygen; (2) end-tidal concentration of remaining sevoflurane 2.15%;⁶ (3) CSI between 40 and 60; (4) having a regular respiratory pattern; and (5) nonresponsive to stimulus such as suctioning.² Gentle suction of oral secretions was allowed in both groups at the discretion of the anesthesiologist responsible for the case after removal of LMA or at the end of surgery in the FM group, and all children breathed 100% oxygen with face mask until they were fully awake and cooperative. Patients were followed up for any advent of complications including cough, apnea (no air movement for more than 10 s), laryngospasm (inspiratory stridor for more than 10 s), bronchospasm, desaturation, vomiting, hypotension (more than 20% decrease in mean arterial pressure lasting for more than 1 min), dysrhythmia, cardiac arrest or death and the data recorded on a sheet. The severity of complications was scored using a previously published scale (Appendix 2).⁴

Position changes required to maintain airway patency were recorded. Parents were contacted by telephone on the 3^rd and 7^th day following discharge to determine any complication including worsening cough, sore throat or re-admission to hospital for respiratory care.

On the basis of a pilot study, we assumed that the incidence of cough (primary outcome) in children with uncomplicated URI anesthetized with LMA would be 30%. Tait et al assumed that by using FM the incidence of complications would become close to that of those children without URI, namely about 5%. Given these two pieces of data, we would need 71 patients in each group to detect such difference (β = 20%, α = 5%). We enrolled 150 children to allow for dropouts. Parametric data were compared using an independent t test. Nonparametric data were analyzed using χ² and, where appropriate, the Wilcoxon and Mann–Whitney U test. Significance was accepted at the 5% level (p < 0.05).

3. Results

One hundred and fifty pediatric patients were enrolled in this study (Fig. 1). They were stratified according to the severity of respiratory tract infection at the preoperative visit. Overall, only 30 patients had moderate URI and 120 had mild URI. Each group of children (mild or moderate) was randomized to receive LMA (n = 75) or FM (n = 75) for anesthesia according to protocol. Three children in the LMA group and two in the FM group were excluded as their plan for ophthalmic intervention had changed. Altogether 73 patients in the FM group and 72 in the LMA group completed the trial. There were no differences between the two groups regarding age, sex, weight, ASA physical status, duration of surgery and severity of URI symptoms (Table 1). There was no significant difference in the prevalence of URI symptoms or the URI onset time (Table 2). The presence of URI was confirmed in all patients on the day of surgery.

The incidences of perioperative complications are depicted in Table 3. There were no difference between the two groups with respect to apnea (7% in the LMA and 5% in the FM group), desaturation (21% in LMA and 20% in FM), laryngospasm (32% in LMA and 37% in FM), bronchospasm (17% in LMA and 14% in FM), readmission (3% in LMA and 4% in FM) and sore throat (18% in LMA and 20% in FM).

In addition there was no occurrence of arrhythmia, cardiac arrest, hypotension or death. We found the incidences of cough and vomiting were statistically higher in the FM group (Table 3). The FM group also needed more changes in position to maintain airway patency.

Twenty patients (33%) experienced laryngospasm; however muscle relaxants were not required and the symptoms resolved spontaneously with positive pressure ventilation. Ninety percent of these respiratory adverse events occurred in patients during emergence from anesthesia.

4. Discussion

There is much bitter controversy in the literature about anesthetizing a child with URI; some studies report that the practice causes only minimal morbidity while others suggest that it increases the risk of perioperative respiratory complications.1, 2, 3, 4, 5, 7 Tait et al assumed that, among children with URI undergoing anesthesia, the incidence of respiratory complications would be greater with the use of LMA than it would be with FM anesthesia, but less than that associated with endotracheal tube anesthesia.1, 2, 3 The assumption regarding more complications with ETT anesthesia has been confirmed in previous studies.1, 2, 3, 4

One issue when debating the pros and cons of different ways of giving anesthesia to children with UTI is the definition of a URI. Although previous studies used specific criteria to define a URI, a retrospective study by Schreiner et al⁸ suggested that these criteria are too stringent and that parental confirmation of a URI was a better predictor of laryngospasm than the use of predetermined criteria.² Therefore, in the present study, diagnosis of a URI was based on the confirmation by parents or guardians in addition to the presenting symptoms noted by the physician.

In our randomized study the severity of URI symptoms in the pediatric patients were stratified. Details of the child’s preoperative URI symptoms were collected by a researcher, so that the anesthesia provider was not fully aware of the symptoms recorded preoperatively.

We can conclude that GA conducted with a laryngeal mask airway can decrease the incidence of postoperative respiratory complications in children with uncomplicated URI as compared to that seen with use of a face mask. The incidence of cough (19% in LMA vs. 42% in FM), vomiting (4% in LMA vs. 12% in FM) and maneuvers to maintain the patency of airway was statistically lower in the LMA group (p < 0.05). However, there was no significant difference with respect to perioperative complications such as apnea, laryngospasm, desaturation, bronchospasm, or with respect to sore throat and readmission. We assumed that secretions of the oral and pharyngeal cavities were effectively cleaned when the LMA was removed.

Previous studies showed discrepant incidences of adverse events in pediatric patients with upper respiratory tract infection.1, 4 We think that the severity of initial symptoms would affect the outcomes; in our study the incidence of cough (primary outcome) was 11/15 (73%) in patients with moderate symptoms who received FM anesthesia compared with 8/15 (53%) in those with moderate symptoms who received LMA anesthesia. That was the rationale to stratify patients by the severity of symptoms.

Thirteen patients in LMA group and 15 in the FM group experienced sore throat, although the difference was not statistically significant. We believe that the greater number of sore throats in the FM group could be a result of deeper suction of the pharyngeal cavity, which was carried out at the discretion of the responsible anesthesiologist. We think that in the LMA group the frequency of suctions for oral secretions was less, and that the suction was not deep enough to affect the posterior parts of pharynx.

This finding concurs with a meta-analysis study that found some advantages of LMA over ETT and FM.⁹ Moreover, in another study by Hoffmann et al on children, fewer adverse airway events were seen with LMA,¹⁰ yet contrary to this, Bordet et al in a prospective study demonstrated more complications with LMA than with FM.¹¹ The present study, which focused on children with mild/moderate URI, showed LMA to be superior.

Children in our study had no co-morbidities and underwent a short noninvasive procedure; therefore the results of this paper cannot be extrapolated to all pediatric patients with URIs. Although topical lidocaine on LMA has been advocated by Schebesta et al¹² in a randomized double-blind study in children with URI, the effect of intravenous medication was not evaluated.1, 2, 3, 10, 11, 12 In our study only intravenous lidocaine was used in both groups to reduce the reactivity of airway.

The potential for bias was minimized by the design of the study; this included stratification of the confounding factors (severity of URI), enrolment of a statistically adequate number of subjects, and the fact that observers had no vested interest in the study’s outcome.

Some short-lived and relatively mild complications were recorded in this study, concurring with previous studies; however, it may be possible to conduct safe GA for elective procedures in children with URI, with meticulous management of symptomatic children and careful attention to the severity of presenting symptoms. Our study indicates that the laryngeal mask airway might be a superior alternative to a facemask, with fewer complications and less need to change the patient’s position.

Acknowledgments

The authors thank Dr Laleh Alim Marvasti (DDS), for her kind contributions in editing this paper, and Ms Leila Mokhtari for her administrative assistance with conducting this study.

Appendix 1.

This sheet was completed by an anesthesiologist, for each patient in our study, during a preoperative visit on the day of surgery. Patients scoring >2 in at least two items were considered to have moderate symptoms of upper respiratory tract infection; otherwise, symptoms were categorized as mild. The sheet was modified from a study designed by Parnis et al.⁴

Do you think that your child has a common cold?

1 – yes 2 – no

When did it start?

______ days ago

Does any one of parents smoke at home?

1 – yes 2 – no

Does the child have any one of the following symptoms?

A: Runny nose

1 – no 2 – sniffing occasionally 3 – continuously running/sniffing, clear 4 – continuously running/sniffing, purulent

B: Nasal congestion

1 – no 2 – difficult breathing through nose 3 – mouth breathing

C: Sneezing

1 – no 2 – occasional 3 – frequent 4 – continuous

D: Cough

1 – no cough 2 – occasional cough 3 – frequent cough 4 – continuous cough

E: Sputum

1 – dry cough 2 – moist cough, no sputum 3 – clear sputum 4 – purulent sputum

Appendix 2.

This form was completed by the responsible anesthesiologist during and after anesthesia. He or she was unaware of the severity of symptoms at the preoperative visit. This form is modified from a study by Parnis et al.⁴

Does the patient have any one of the following symptoms?

Please specify when it occurred (during induction, maintenance, recovery, or at home)

A: Cough

1 – no 2 – yes, not troublesome 3 – yes, interferes with ventilation/oxygenation

B: Bronchospasm

1 – none 2 – expiratory or inspiratory wheeze 3 – inspiratory and expiratory wheeze 4 – difficult to ventilate

C: Apnea (no air movement for > 10 s)

1 – yes 2 – no

D: Laryngospasm (inspiratory stridor for > 10 s)

1 – yes 2 – no

E: Desaturation

1 – Saturation remained > 95%

2 – Saturation 90– 5% resolved spontaneously

3 – Saturation 90– 95% required treatment

4 – Saturation 90–95% despite treatment

5 – Saturation < 90% resolved spontaneously

6 – Saturation < 90% resolved by CPAP (continuous positive airway pressure)

7 – Saturation < 90% despite intervention

F: Vomiting

1 – no 2 – once 3 – more than once

G: Sore throat

1 – no 2 – difficult swallowing 3 – unable to swallow

H: Hypotension (mean arterial pressure decreased > 20 % from base for more than 1 min)

1 – yes 2 – no

I: Arrhythmia (any arrhythmia that needed medication to treat)

1 – yes 2 – no

J: Cardiac arrest

1 – yes 2 – no

K: Death

1 – yes 2 – no

Did the symptoms increase, the night after surgery?

1 – yes 2 – no

Was the patient readmitted to hospital for respiratory problems within 1 week of anesthesia?

1 – yes 2 – no